/
modele.py
123 lines (84 loc) · 2.47 KB
/
modele.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
__author__ = 'Jordan Guerin'
import numpy
import math
import random
from data import dataset
data = dataset()
data.load()
x_train = data.points[0:80]
#x_valid = data.points[60:80]
x_test = data.points[80:]
def shuffle(pts, nb=100):
i1 = random.randint(0, len(pts)-1)
i2 = random.randint(0, len(pts)-1)
pt1 = pts[i1]
pt2 = pts[i2]
pts[i2] = pt1
pts[i1] = pt2
return pts
def entrainerModele(pts, deg=0):
a_x = []
a_y = []
for p in pts:
a_x.append(p.x)
a_y.append(p.y)
w = numpy.polynomial.polynomial.polyfit(a_x, a_y, deg)
return w
def evaluerModele(p, x_sel):
n = float(len(x_sel))
sum = 0.0000000
for pt in x_sel:
p_y = numpy.polyval(p, pt.x)
#print p_y
sum += math.pow(pt.y - p_y, 2)
return sum/n
if __name__ == '__main__':
print "Influence du nombre de donnee"
x_sel = shuffle(x_train)[0:10]
w = entrainerModele(x_sel, 3)
poly = numpy.poly1d(w)
err = evaluerModele(poly, x_sel)
print "Xsel : %.06f" % err
fp = open("test.out", 'w')
for i in range(10, 81, 5):
x_sel = shuffle(data.points)
x_sel = x_train[0:i]
#x_test = x_train[len(x_train)-i:]
w = entrainerModele(x_train, 3)
poly = numpy.poly1d(w)
err = evaluerModele(poly, x_sel)
print "E_emp[%d] : %.06f" % (i, err)
err2 = evaluerModele(poly, x_test)
print "E_gen[%d] : %.06f" % (i, err2)
fp.write(("%d\t%0.6f\t%0.6f\n" % (i, err, err2)).replace('.', ','))
fp.close()
print ""
print "Influence de l'ordre de regression"
x_train = data.points[0:60]
x_valid = data.points[60:80]
x_test = data.points[80:]
e_emp = []
e_gen = []
degs = []
fp = open('test2.out', 'w')
for deg in range(21):
degs.append(deg)
w = entrainerModele(x_train, deg)
poly = numpy.poly1d(w)
err = evaluerModele(poly, x_valid)
print "E_emp[%d] : %0.6f" % (deg, err)
e_emp.append(err)
err2 = evaluerModele(poly, x_test)
print "E_gen[%d] : %0.6f" % (deg, err2)
e_gen.append(err2)
fp.write(("%d\t%0.6f\t%0.6f\n" % (deg, err, err2)).replace('.', ','))
fp.close()
#import matplotlib.pyplot as plot
#plot.plot(degs, e_emp, 'b.', degs, e_gen, "g.")
#plot.show()
#test p*
#meilleur ordre est 1
w = entrainerModele(x_train, 1)
poly = numpy.poly1d(w)
err = evaluerModele(poly, x_test)
print err