描述过几天再写，先放代码

#!/usr/bin/python
# -*- coding: utf-8 -*-


import random
import matplotlib.pyplot as plt
from sklearn.decomposition import PCA

list_all = []
cluster1_set = []
cluster2_set = []
cluster3_set = []
list_data = []   # 数据


def read_data_set():   #处理数据集
    filename = open('iris.dat', 'r', encoding='utf-8')
    lines = filename.readlines()
    for line in lines:
        if line[0] != '@':   # 去除数据集前面的描述语句
            list_temp = line.strip().split(', ')
            list_float = []
            for i in list_temp[0:-1]:
                list_float.append(float(i))   #string转化为float
            list_float.append(list_temp[-1])
            list_all.append(list_float)
    pass

# 计算每一迭代的新中心点
def aver_list(list_temp0):
    sum_0 = 0
    sum_1 = 0
    sum_2 = 0
    sum_3 = 0
    for a1 in list_temp0:
        sum_0 += a1[0]
        sum_1 += a1[1]
        sum_2 += a1[2]
        sum_3 += a1[3]
    return [sum_0 / len(list_temp0), sum_1 / len(list_temp0), sum_2 / len(list_temp0), sum_3 / len(list_temp0)]

def k_means():
    #  找初始的三个簇中心
    a = random.randint(0, 49)
    b = random.randint(50, 99)
    c = random.randint(100, 149)
    cluster1 = list_all[a][0:-1]
    cluster2 = list_all[b][0:-1]
    cluster3 = list_all[c][0:-1]
    print(cluster1)
    print(cluster2)
    print(cluster3)

    count = 0
    while True:
        for list_data in list_all[:]:
            #   欧氏距离
            sum1 = ((cluster1[0] - list_data[0]) ** 2 + (cluster1[1] - list_data[1]) ** 2 \
                   + (cluster1[2] - list_data[2]) ** 2 + (cluster1[3] - list_data[3]) ** 2) ** 0.5
            sum2 = ((cluster2[0] - list_data[0]) ** 2 + (cluster2[1] - list_data[1]) ** 2 \
                   + (cluster2[2] - list_data[2]) ** 2 + (cluster2[3] - list_data[3]) ** 2) ** 0.5
            sum3 = ((cluster3[0] - list_data[0]) ** 2 + (cluster3[1] - list_data[1]) ** 2 \
                   + (cluster3[2] - list_data[2]) ** 2 + (cluster3[3] - list_data[3]) ** 2) ** 0.5
            if sum1 == min(sum1, sum2, sum3):
                cluster1_set.append(list_data)
            elif sum2 == min(sum1, sum2, sum3):
                cluster2_set.append(list_data)
            elif sum3 == min(sum1, sum2, sum3):
                cluster3_set.append(list_data)
        #   中心点
        average1 = aver_list(cluster1_set)
        average2 = aver_list(cluster2_set)
        average3 = aver_list(cluster3_set)
        #   计算中心点和上一任中心点之间的距离，判断是否停止迭代
        sum1 = ((cluster1[0] - average1[0]) ** 2 + (cluster1[1] - average1[1]) ** 2 \
                + (cluster1[2] - average1[2]) ** 2 + (cluster1[3] - average1[3]) ** 2) ** 0.5
        sum2 = ((cluster2[0] - average2[0]) ** 2 + (cluster2[1] - average2[1]) ** 2 \
                + (cluster2[2] - average2[2]) ** 2 + (cluster2[3] - average2[3]) ** 2) ** 0.5
        sum3 = ((cluster3[0] - average3[0]) ** 2 + (cluster3[1] - average3[1]) ** 2 \
                + (cluster3[2] - average3[2]) ** 2 + (cluster3[3] - average3[3]) ** 2) ** 0.5
        if (sum1 < 1e-12) & (sum2 < 1e-12) & (sum3 < 1e-12):   # 临界值
            break
        else:
            #   中心点成为新的簇点
            cluster1 = average1
            cluster2 = average2
            cluster3 = average3
            count += 1
            print("第" +str(count) + "次迭代")
            print(cluster1)
            print(cluster2)
            print(cluster3)
        # 达到一定的迭代次数，终止
        if count > 100000:  # 迭代次数
            break

        #   清空 簇list
        cluster1_set.clear()
        cluster2_set.clear()
        cluster3_set.clear()
    print("--------------------------")
    print(cluster1)
    print(cluster2)
    print(cluster3)

    pass

# 计算分类正确率
def sort_correctly():
    sort1 = 0
    sort2 = 0
    sort3 = 0
    sum = 0
    # 簇1的正确数
    for list_wrong1 in cluster1_set[:]:
        if list_wrong1[-1] == "Iris-setosa":
            sort1 += 1
        elif list_wrong1[-1] == "Iris-versicolor":
            sort2 += 1
        else:
            sort3 += 1
    sum += max(sort1, sort2, sort3)
    sort1 = 0
    sort2 = 0
    sort3 = 0
    # 簇2的正确数
    for list_wrong2 in cluster2_set[:]:
        if list_wrong2[-1] == "Iris-setosa":
            sort1 += 1
        elif list_wrong2[-1] == "Iris-versicolor":
            sort2 += 1
        else:
            sort3 += 1
    sum += max(sort1, sort2, sort3)
    sort1 = 0
    sort2 = 0
    sort3 = 0
    # 簇3的正确数
    for list_wrong3 in cluster3_set[:]:
        if list_wrong3[-1] == "Iris-setosa":
            sort1 += 1
        elif list_wrong3[-1] == "Iris-versicolor":
            sort2 += 1
        else:
            sort3 += 1
    sum += max(sort1, sort2, sort3)

    num_right = "%.2f%%" % ((sum / len(list_all)) * 100)
    print("聚类分析的正确率为" + str(num_right))
    pass

def dimension_reduction():
    #数据和属性一一对应
    for i in range(len(list_all)):
        list_data.append(list_all[i][0:-1])
    pca = PCA(2)
    pca.fit(list_data)
    low_list_data = pca.transform(list_data)
    return low_list_data
    pass


def print_scatter():
    low_data = dimension_reduction()
    cluster1_set_x = []
    cluster1_set_y = []
    cluster2_set_x = []
    cluster2_set_y = []
    cluster3_set_x = []
    cluster3_set_y = []
    for i in range(len(list_all)):
        if list_all[i] in cluster1_set:
            cluster1_set_x.append(low_data[i][0])
            cluster1_set_y.append(low_data[i][1])
        elif list_all[i] in cluster2_set:
            cluster2_set_x.append(low_data[i][0])
            cluster2_set_y.append(low_data[i][1])
        elif list_all[i] in cluster3_set:
            cluster3_set_x.append(low_data[i][0])
            cluster3_set_y.append(low_data[i][1])
    plt.scatter(cluster1_set_x, cluster1_set_y, c='red', marker='+')
    plt.scatter(cluster2_set_x, cluster2_set_y, c='green', marker='*')
    plt.scatter(cluster3_set_x, cluster3_set_y, c='blue', marker='s')
    plt.show()
    pass


if __name__ == "__main__":
    read_data_set()
    k_means()
    sort_correctly()
    print_scatter()

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