KNN代码实现

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
# coding = utf8
# 加载knn
from sklearn import neighbors
import numpy as np

def creat_datasets():
    datasets = np.array([[8, 4, 2],
                         [7, 1, 1],
                         [1, 4, 4],
                         [3, 0, 5],
                         [9, 4, 2],
                         [7, 0, 1],
                         [1, 5, 4],
                         [4, 0, 5]
                        ])
    label = [0,0,1,1,0,0,1,1]
    return datasets, label


def knn_sklearn_predict():
    #调用KNN
    knn = neighbors.KNeighborsClassifier()
    #加载数据集
    dataset,label = creat_datasets()
    #传入数据
    knn.fit(dataset,label)
    #预测
    result = knn.predict([[2,4,0]])
    print(result)
    print('非常热'if result[0]==0 else '一般热')
    return result


if __name__ == '__main__':
    knn_sklearn_predict()
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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
import numpy as np
import matplotlib
import matplotlib.pyplot as plt
import math


def creat_datasets():
    datasets = np.array([[8, 4, 2],
                         [7, 1, 1],
                         [1, 4, 4],
                         [3, 0, 5]])
    label = ['very hot', 'very hot', 'hot', 'hot']
    return datasets, label


def analyze_data_plot(x, y):
    fig = plt.figure()
    ax = fig.add_subplot(111)
    ax.scatter(x, y)

    plt.title('test title')
    plt.xlabel('icecream')
    plt.ylabel('drink water')
    plt.show()


def computeEuclideanDistance(x1, y1, x2, y2):
    return math.sqrt(math.pow((x1 - x2), 2) + math.pow((y1 - y2), 2))
    #二维

    # 多维
def computeEuclideanDistance2(arr1, arr2, length):
    d = 0;
    for i in range(length):
        d += math.pow(arr1[i] - arr2[i], 2)
    return math.sqrt(d)

    #矩阵
def computeEuclideanDistance3(newV,dataSet):
    #获取形状
    rowSize, colSize = dataSet.shape
    #各特征向量间做差
    diffMat = np.tile(newV,(rowSize,1))-dataSet
    #计算欧式距离的步骤平方
    sq2 = diffMat**2
    #求合并开方
    return (sq2.sum(axis=1)**0.5)

def knn_Classifier(newV, datasets, label, topk):
    import operator
    # 1.计算新数据和样本数据之间的距离
    dist = computeEuclideanDistance3(newV,datasets)
    # 2.根据距离进行排序
    sortedDistIndexs = dist.argsort(axis=0)
    #print(sortedDistIndexs)
    # 3.针对topk统计类别进行排序。
    disCount = {}
    for i in range(topk):
        votelabel = label[sortedDistIndexs[i]]
        #print(sortedDistIndexs[i], votelabel)
        disCount[votelabel] = disCount.get(votelabel,0)+1
        #可以拆成disCount.get(votelabel,0)->disCount[votelabel]=0
        #disCount[votelabel]=disCount[votelabel]+1
        # 这里着重讲一下第4行代码：cou[i] = cou.get(i, 0) + 1
        # 第行代码从逻辑上讲执行了两次，而这两次里get语句每次执行时的功能是不一样的：
        #
        # 第一次：cou[i] = cou.get(i, 0) + 1（i =‘aa’）
        # 此时get语句的功能为赋初值，即把键’aa’的初值置为0然后加1
        #
        # 第二次：cou[i] = cou.get(i, 0) + 1（i =‘aa’）
        # 因get语句已经作为赋值语句出现过一次了，因此此时再执行这条语句时，赋值功能已经无效了，也就是get语句里第二个参数对’aa’这个键已经无效了，此时get语句的功能为文章内的第一大部分所介绍的功能，所以这条语句此时可等价为
        # cou[i] = cou.get(i) + 1
        #
        # 即
        # cou[i] = cou[i] + 1

        #print(disCount)
        #从大到小排序
    sortedDisCount = sorted(disCount.items(),key=operator.itemgetter(1),reverse=True)
    #print('result:',sortedDisCount[0][0])
    return sortedDisCount[0][0]

def predict():
    dataset,label = creat_datasets()
    newV=[2,4,4]
    a = float(input())
    b = float(input())
    c = float(input())
    newV2=[a,b,c]
    result = knn_Classifier(newV2,dataset,label,3)
    print(result)

if __name__ == '__main__':
    # datasets, label = creat_datasets()
    # print(datasets, '\n', label)
    # analyze_data_plot(datasets[:, 0], datasets[:, 1])
    #
    # # 计算欧式距离
    # d = computeEuclideanDistance(2, 4, 8, 2)
    # print('eu1:',d)
    # d2 = computeEuclideanDistance2([2,4,4],[7,1,1],3)
    # print('eu2:',d2)
    # d3 = computeEuclideanDistance3([2,4,4],datasets)
    # print('eu3:', d3)
    # # knn
    # newV = [2, 4, 0]
    # knn=knn_Classifier(newV, datasets, label, 3)
    # print('knn_result', knn)
    #
    # vecs = np.array([[2,4,4],[3,0,0],[5,7,2]])
    # for v in vecs:
    #     print('3 knn_result:',knn_Classifier(v,datasets,label,3))
    predict()import numpy as np
    import matplotlib
    import matplotlib.pyplot as plt
    import math


    def creat_datasets():
    datasets = np.array([[8, 4, 2],
    [7, 1, 1],
    [1, 4, 4],
    [3, 0, 5]])
    label = ['very hot', 'very hot', 'hot', 'hot']
    return datasets, label


    def analyze_data_plot(x, y):
    fig = plt.figure()
    ax = fig.add_subplot(111)
    ax.scatter(x, y)

    plt.title('test title')
    plt.xlabel('icecream')
    plt.ylabel('drink water')
    plt.show()


    def computeEuclideanDistance(x1, y1, x2, y2):
    return math.sqrt(math.pow((x1 - x2), 2) + math.pow((y1 - y2), 2))
    #二维

    # 多维
    def computeEuclideanDistance2(arr1, arr2, length):
    d = 0;
    for i in range(length):
    d += math.pow(arr1[i] - arr2[i], 2)
    return math.sqrt(d)

    #矩阵
    def computeEuclideanDistance3(newV,dataSet):
    #获取形状
    rowSize, colSize = dataSet.shape
    #各特征向量间做差
    diffMat = np.tile(newV,(rowSize,1))-dataSet
    #计算欧式距离的步骤平方
    sq2 = diffMat**2
    #求合并开方
    return (sq2.sum(axis=1)**0.5)

    def knn_Classifier(newV, datasets, label, topk):
    import operator
    # 1.计算新数据和样本数据之间的距离
    dist = computeEuclideanDistance3(newV,datasets)
    # 2.根据距离进行排序
    sortedDistIndexs = dist.argsort(axis=0)
    #print(sortedDistIndexs)
    # 3.针对topk统计类别进行排序。
    disCount = {}
    for i in range(topk):
    votelabel = label[sortedDistIndexs[i]]
    #print(sortedDistIndexs[i], votelabel)
    disCount[votelabel] = disCount.get(votelabel,0)+1
    #可以拆成disCount.get(votelabel,0)->disCount[votelabel]=0
    #disCount[votelabel]=disCount[votelabel]+1
    # 这里着重讲一下第4行代码：cou[i] = cou.get(i, 0) + 1
    # 第行代码从逻辑上讲执行了两次，而这两次里get语句每次执行时的功能是不一样的：
    #
    # 第一次：cou[i] = cou.get(i, 0) + 1（i =‘aa’）
    # 此时get语句的功能为赋初值，即把键’aa’的初值置为0然后加1
    #
    # 第二次：cou[i] = cou.get(i, 0) + 1（i =‘aa’）
    # 因get语句已经作为赋值语句出现过一次了，因此此时再执行这条语句时，赋值功能已经无效了，也就是get语句里第二个参数对’aa’这个键已经无效了，此时get语句的功能为文章内的第一大部分所介绍的功能，所以这条语句此时可等价为
    # cou[i] = cou.get(i) + 1
    #
    # 即
    # cou[i] = cou[i] + 1

    #print(disCount)
    #从大到小排序
    sortedDisCount = sorted(disCount.items(),key=operator.itemgetter(1),reverse=True)
    #print('result:',sortedDisCount[0][0])
    return sortedDisCount[0][0]

    def predict():
    dataset,label = creat_datasets()
    newV=[2,4,4]
    a = float(input())
    b = float(input())
    c = float(input())
    newV2=[a,b,c]
    result = knn_Classifier(newV2,dataset,label,3)
    print(result)

    if __name__ == '__main__':
    # datasets, label = creat_datasets()
    # print(datasets, '\n', label)
    # analyze_data_plot(datasets[:, 0], datasets[:, 1])
    #
    # # 计算欧式距离
    # d = computeEuclideanDistance(2, 4, 8, 2)
    # print('eu1:',d)
    # d2 = computeEuclideanDistance2([2,4,4],[7,1,1],3)
    # print('eu2:',d2)
    # d3 = computeEuclideanDistance3([2,4,4],datasets)
    # print('eu3:', d3)
    # # knn
    # newV = [2, 4, 0]
    # knn=knn_Classifier(newV, datasets, label, 3)
    # print('knn_result', knn)
    #
    # vecs = np.array([[2,4,4],[3,0,0],[5,7,2]])
    # for v in vecs:
    #     print('3 knn_result:',knn_Classifier(v,datasets,label,3))
    predict()