ml/Breading_Room.py at master · code76kb/ml · GitHub

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

import numpy as np
from Full_Snake import Full_Snake
import random


class Breading_Room:

    OFFSPRINGS = 2

    def __init__(self, snak_len):
        self.snak_len = snak_len
        print("Breading_Room is reqdy")


    def make_love(self,mates_array):
        print("MAking love total MAtes:",len(mates_array))
        male_Snaks = []
        female_Snaks= []
        topHalfs = []
        lowHalf = []
        newBread = []
        sortedByFitness = sorted(mates_array,key=lambda x:x.fitness, reverse=True)
        topHalfs = sortedByFitness[:len(mates_array)/2]
        lowHalf = sortedByFitness[len(mates_array)/2:]

        print("topHalfs size :",len(topHalfs))

        # for i in xrange(0,len(topHalfs)):
        #     print("\nSorted by fitness",sortedByFitness[i].fitness)
        #     if i%2 == 0:
        #         male_Snaks.append(topHalfs[abs(len(topHalfs)-i)])
        #     else:
        #         female_Snaks.append(topHalfs[abs(len(topHalfs)-i)])
        female_Snaks = topHalfs
        male_Snaks = topHalfs[::-1]

        print("Total Femals :",len(female_Snaks))
        print("Total Males :",len(male_Snaks))
        pair = len(male_Snaks) if(len(female_Snaks)>len(male_Snaks)) else len(female_Snaks)

        for j in xrange(0,len(topHalfs)):
            weight_H_female = female_Snaks[j].nural_Net.weight_matrix_hidden
            weight_H_male = male_Snaks[j].nural_Net.weight_matrix_hidden
            weight_O_female = female_Snaks[j].nural_Net.weight_matrix_output
            weight_O_male = male_Snaks[j].nural_Net.weight_matrix_output

            # print("\n Femals H:",weight_H_female)
            # print("\n Femals O:",weight_O_female)
            # print("\n Mals H:",weight_H_male)
            # print("\n Mals O:",weight_O_male)

            # print("\n Split::",np.array_split(weight_H_female,2))

            w_h_F_UpearHELF = np.array_split(weight_H_female,2)[0] #4*2
            w_h_F_LowerHELF = np.array_split(weight_H_female,2)[1] #4*3

            w_h_M_UpearHELF = np.array_split(weight_H_male,2)[0] #4*2
            w_h_M_LowerHELF = np.array_split(weight_H_male,2)[1] #4*3


            w_o_F_UpearHELF = np.array_split(weight_O_female,2)[0] #5*2
            w_o_F_LowerHELF = np.array_split(weight_O_female,2)[1] #5*2

            w_o_M_UpearHELF = np.array_split(weight_O_female,2)[0] #5*2
            w_o_M_LowerHELF = np.array_split(weight_O_female,2)[1] #5*2

            #Boy
            w_h_B = np.vstack((w_h_M_UpearHELF, w_h_F_LowerHELF))
            w_o_B = np.vstack((w_o_M_UpearHELF, w_o_F_LowerHELF))

            #mutation
            r = (w_h_B.shape[0])-1
            c = (w_h_B.shape[1])-1

            m1 =  1 / (1 + np.e ** -w_h_B[random.randint(0,r),random.randint(0,c)])
            # print('Mutation b1 by:',m1)
            w_h_B[random.randint(0,r),random.randint(0,c)] =  m1

            r = (w_o_B.shape[0])-1
            c = (w_o_B.shape[1])-1
            m2 = 1 / (1 + np.e ** -w_o_B[random.randint(0,r),random.randint(0,c)])
            # print('Mutation b2 by:',m2)
            w_o_B[random.randint(0,r),random.randint(0,c)] =  m2


            newSnak = Full_Snake(self.snak_len)
            newSnak.nural_Net.weight_matrix_hidden = w_h_B
            newSnak.nural_Net.weight_matrix_output = w_o_B

            newBread.append(newSnak)

            #Girl
            w_h_G = np.vstack((w_h_M_UpearHELF, w_h_F_LowerHELF))
            w_o_G = np.vstack((w_o_M_UpearHELF, w_o_F_LowerHELF))

            #mutation
            r = (w_h_G.shape[0])-1
            c = (w_h_G.shape[1])-1
            m3 = 1 / (1 + np.e ** -w_h_G[random.randint(0,r),random.randint(0,c)])
            # print('Mutation g1 by:',m3)
            w_h_G[random.randint(0,r),random.randint(0,c)] =  m3

            r = (w_o_G.shape[0])-1
            c = (w_o_G.shape[1])-1
            m4 =  1 / (1 + np.e ** -w_o_G[random.randint(0,r),random.randint(0,c)])
            # print('Mutation G2 by:',m4)
            w_o_G[random.randint(0,r),random.randint(0,c)] = m4

            newSnak = Full_Snake(self.snak_len)
            newSnak.nural_Net.weight_matrix_hidden = w_h_G
            newSnak.nural_Net.weight_matrix_output = w_o_G

            # print("\n BOY H:",w_h_B)
            # print("\n BOY O:",w_o_B)
            # print("\n GIRL H:",w_h_G)
            # print("\n GIRL O:",w_o_G)

            newBread.append(newSnak)

        print("new Bread popolation :",len(newBread))
        return newBread