Statistical-Learning-Algorithm/SVM.py at main · aky15/Statistical-Learning-Algorithm · 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
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
import numpy as np
import pandas as pd
from sklearn.datasets import load_iris
from sklearn.model_selection import  train_test_split
import matplotlib.pyplot as plt
import matplotlib as mpl
mpl.use('tkagg')
def create_data():
    iris = load_iris()
    df = pd.DataFrame(iris.data, columns=iris.feature_names)
    df['label'] = iris.target
    df.columns = [
        'sepal length', 'sepal width', 'petal length', 'petal width', 'label'
    ]
    data = np.array(df.iloc[:100, [0, 1, -1]])
    for i in range(len(data)):
        if data[i, -1] == 0:
            data[i, -1] = -1
    return data[:, :2], data[:, -1]

X, y = create_data()
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25)

class SVM(object):
    def __init__(self, kernel, X_train, y_train, C=1, p=2, sigma=1):
        self.kernel = kernel
        self.X_train = X_train
        self.y_train = y_train
        self.m = X_train.shape[0]
        self.n = X_train.shape[1]
        self.alpha = np.random.random(self.m)
        self.K = self.ComputeKernel(kernel, X_train, p, sigma)
        self.C = C
        self.b = 0
        self.p = p
        self.sigma = sigma
        self.E = [ self._E(i) for i in range(self.m)]

    def ComputeKernel(self, kernel, X_train, p, sigma):
        if kernel == "linear":
            return np.dot(X_train, X_train.transpose())
        elif kernel == "poly":
            return (np.dot(X_train, X_train.transpose())+1) ** p
        elif kernel == "gaussian":
            G = np.dot(X_train, X_train.transpose())
            H = np.tile(np.diag(G), (self.m,1))
            return np.exp(-(H + H.T - 2*G)/(2*sigma**2))
        else:
            raise ValueError("kernel must be in linear, poly or gaussian")

    def _E(self, i):
        return self._g(i) - self.y_train[i]

    def _g(self, i):
        return np.dot(self.alpha * self.y_train, self.K[i][:]) + self.b

    def _KKT(self, i):
        y_g = self._g(i) * self.y_train[i]
        if self.alpha[i] == 0:
            return y_g >= 1
        elif 0 < self.alpha[i] < self.C:
            return y_g == 1
        else:
            return y_g <= 1

    def select(self):
        index_list = [i for i in range(self.m) if 0 < self.alpha[i] < self.C]
        non_boundary = [i for i in range(self.m) if i not in index_list]
        index_list.extend(non_boundary)
        for i in index_list:
            if self._KKT(i):
                continue
            E1 = self.E[i]
            if E1 >= 0:
                j = min(range(self.m), key=lambda x: self.E[x])
            else:
                j = max(range(self.m), key=lambda x: self.E[x])
            return i, j

    def _clip(self, val, L, H):
        if val<L:
            return L
        elif val>H:
            return H
        return val

    def train(self, max_iter=1):
        for _ in range(max_iter):
            i1, i2 = self.select()
            if self.y_train[i1] == self.y_train[i2]:
                L = max(0, self.alpha[i1] + self.alpha[i2] - self.C)
                H = min(self.C, self.alpha[i1] + self.alpha[i2])
            else:
                L = max(0, self.alpha[i2] - self.alpha[i1])
                H = min(self.C, self.C + self.alpha[i2] - self.alpha[i1])

            E1 = self.E[i1]
            E2 = self.E[i2]

            eta = self.K[i1][i1] + self.K[i2][i2] - 2 * self.K[i1][i2]
            if eta <= 0:
                continue
            alpha2_new_unc = self.alpha[i2] + self.y_train[i2] * (E1 - E2) / eta
            alpha2_new = self._clip(alpha2_new_unc, L, H)
            alpha1_new = self.alpha[i1] + self.y_train[i1] * self.y_train[i2] * (self.alpha[i2] - alpha2_new)

            b1_new = -E1 - self.y_train[i1] * self.K[i1][i2] * (
                alpha1_new - self.alpha[i1]) - self.y_train[i2] * self.K[i2][i1] * (alpha2_new - self.alpha[i2]) + self.b
            b2_new = -E2 - self.y_train[i1] * self.K[i1][i2] * (
                alpha1_new - self.alpha[i1]) - self.y_train[i2] * self.K[i2][i1] * (alpha2_new - self.alpha[i2]) + self.b

            if 0 < alpha1_new < self.C:
                b_new = b1_new
            elif 0 < alpha2_new < self.C:
                b_new = b2_new
            else:
                b_new = (b1_new + b2_new) / 2

            self.alpha[i1] = alpha1_new
            self.alpha[i2] = alpha2_new
            self.b = b_new

            self.E[i1] = self._E(i1)
            self.E[i2] = self._E(i2)

    def predict(self, x):
        if self.kernel == "linear":
            r = np.dot((self.alpha*self.y_train).transpose(),np.dot(self.X_train,x.transpose())) + self.b
        elif self.kernel == "poly":
            r = np.dot((self.alpha*self.y_train).transpose(),(np.dot(self.X_train,x.transpose())+1)**self.p) + self.b
        elif self.kernel == "gaussian":
            d = 0
            for i in range(self.n):
                d += (self.X_train - x)[:,i] ** 2
            d = d/(2*self.sigma**2)
            d = np.exp(-d)
            r = np.dot((self.alpha*self.y_train).transpose(),d) + self.b
        return 1 if r>0 else -1

    def score(self, X_test, y_test):
        right_count = 0
        for i in range(len(X_test)):
            result = self.predict(X_test[i])
            if result == y_test[i]:
                right_count += 1
        return right_count / len(X_test)


res = []
for i in range(10):
    svm = SVM("gaussian", X_train, y_train)
    svm.train(300)
    res.append(svm.score(X_test, y_test))
print(np.mean(res))