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
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
| #!/usr/bin/env python
# -*- coding: utf-8 -*-
# Created by PyCharm
# @author : mystic
# @date : 2017/11/13 20:39
from numpy import *
import operator
import matplotlib.pyplot as plt
from os import listdir
def create_data_set():
group = array([[1.0, 1.1], [1.0, 1.0], [0, 1], [0, 0.1]])
labels = ['A', 'C', 'B', 'D']
return group, labels
def classify(in_x, data_set, labels, k):
"""
分类器
:param in_x: 用于分类的输入向量
:param data_set: 输入的训练样本集
:param labels: 标签向量
:param k: 用于选择最近邻居的数目
:return:
"""
# 获取data_set的第一维长度
data_set_size = data_set.shape[0]
# 分别计算输入向量与data_set集合中各点的向量差,并存入数组中
diff_arr = tile(in_x, (data_set_size, 1)) - data_set
# 平方
sq_diff_arr = diff_arr ** 2
# 求平方和
sq_distinces = sq_diff_arr.sum(axis=1)
# 开根,得各点与输入向量的距离值集合
distinces = sq_distinces ** 0.5
# 排序,升序(返回结果为索引,如[17,23,1,0],排序后返回[3,2,0,1])
sorted_dist_indices = distinces.argsort()
# print('最近的点:%s' % labels[sorted_dist_indices[0]])
# 存储最近的k个点
class_count = {}
for i in range(k):
vote_label = labels[sorted_dist_indices[i]]
class_count[vote_label] = class_count.get(vote_label, 0) + 1
# print(class_count)
# 根据字典class_count的value进行降序排列
# 在最近点案例中,value都是1,下面的排序等于没做
sorted_class_count = sorted(class_count.items(),
key=operator.itemgetter(1), reverse=True)
# print(sorted_class_count)
return sorted_class_count[0][0]
def file2array(filename):
# 获取文件行数
fr = open(filename)
array_lines = fr.readlines()
amount = len(array_lines)
# 创建第一维长度为amount,第二维长度为3的零填充数组
# 因为我们需要存储的数据第二维长度为3,故我们设置固定长度3
return_array = zeros((amount, 3))
# 创建空list
class_label_vector = []
index = 0
for line in array_lines:
# strip([chars]) 去除头尾的字符,默认去除空格字符
line = line.strip()
list_from_line = line.split('\t')
# 获取前三个元素,赋值给return_array(这里采用了多维切片)
return_array[index, :] = list_from_line[0:3]
# 负索引从后获取,-1获取最后一个元素
class_label_vector.append(int(list_from_line[-1]))
index += 1
# 返回tuple
return return_array, class_label_vector
def show_data_in_chart():
plt.rcParams['font.sans-serif'] = ['SimHei'] # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False # 用来正常显示负号
dating_data_arr, dating_labels = file2array('resource/datingTestSet2.txt')
# new figure
fig = plt.figure()
# 在1行1列,第一个子图作画;如233,在2行3列共有6个子图的图中,在第3个子图中作画
ax = fig.add_subplot(111)
# 设置标题
ax.set_title('Appointment Data Analysis')
# Helen提供的数据,三列分别是:每年获得的飞行常客里程数,玩视频游戏所耗时间百分比,每周消费的冰淇淋公升数
# 1:not like 2:general like 3:very like
# 因为我们最后显示的是第一列和第二列数据,故如下设置备注信息
plt.xlabel('每年获得的飞行常客里程数')
plt.ylabel('玩视频游戏所耗时间百分比')
# ax.scatter(dating_data_arr[:, 1], dating_data_arr[:, 2])
# ax.scatter(dating_data_arr[:, 1], dating_data_arr[:, 2], 15.0*array(dating_labels), 15.0*array(dating_labels))
# 第一列与第二列的数据,显示效果更优
# ax.scatter(dating_data_arr[:, 0], dating_data_arr[:, 1])
# 这种方式没有图例,难以区分
# ax.scatter(dating_data_arr[:, 0], dating_data_arr[:, 1], 15.0 * array(dating_labels), 15.0 * array(dating_labels))
# 添加图例
length = dating_data_arr.shape[0]
x1 = []
y1 = []
x2 = []
y2 = []
x3 = []
y3 = []
for i in range(length):
if dating_labels[i] == 1:
x1.append(dating_data_arr[i, 0])
y1.append(dating_data_arr[i, 1])
elif dating_labels[i] == 2:
x2.append(dating_data_arr[i, 0])
y2.append(dating_data_arr[i, 1])
else:
x3.append(dating_data_arr[i, 0])
y3.append(dating_data_arr[i, 1])
type1 = ax.scatter(x1, y1, c='red')
type2 = ax.scatter(x2, y2, c='green')
type3 = ax.scatter(x3, y3, c='blue')
ax.legend([type1, type2, type3], ["not like", "general like", "very like"], loc=2)
plt.show()
def auto_norm(data_set):
"""
归一化特征值:自动将数据集转化为0到1区间内的值
由于里程数远远大于其他特征值,对结果影响过大
而Helen认为三者同等重要,故采用归一化处理
:param data_set:
:return:
"""
# 取第一维度的最小值
"""
>>> sh = array([
[[1, 1],[8, 18],[100, 3],[2, 4]],
[[1, 90],[21, 2],[11, 3],[19, 4]]
])
>>> shape(sh)
(2,4,2)
>>> sh.max()
100
>>> sh.min()
1
>>> sh.max(0)
array([[ 1, 90],
[ 21, 18],
[100, 3],
[ 19, 4]])
>>> sh.min(0)
array([[ 1, 1],
[ 8, 2],
[11, 3],
[ 2, 4]])
"""
min_vals = data_set.min(0)
max_vals = data_set.max(0)
ranges = max_vals - min_vals
# 创建一个与data_set各维长度均相同的零填充数组
# norm_data_set = zeros(shape(data_set))
length = data_set.shape[0]
norm_data_set = data_set - tile(min_vals, (length, 1))
norm_data_set = norm_data_set / tile(ranges, (length, 1))
return norm_data_set, ranges, min_vals
def dating_class_test():
"""
测试代码
:return:
"""
# 用于测试的数据,占总数据的百分比
# [如:已有100条数据,其中90条作为样本训练数据,剩余10条用于测试算法,检测算法的正确率]
test_ratio = 0.10
dating_data_arr, dating_labels = file2array('resource/datingTestSet2.txt')
norm_arr, ranges, min_vals = auto_norm(dating_data_arr)
length = norm_arr.shape[0]
# 测试数据总数
num_test_data = int(length * test_ratio)
error_count = 0.0
for i in range(num_test_data):
classifier_result = classify(norm_arr[i, :], norm_arr[num_test_data:length, :],
dating_labels[num_test_data:length], 3)
print('The classifier came back with: %d, the real answer is: %d'
% (classifier_result, dating_labels[i]))
# 如果分类器结果和真实数据,不同;error_count+1
if classifier_result != dating_labels[i]:
error_count += 1.0
print('The total error rate is: %f' % (error_count / num_test_data))
def classify_person():
"""
预测函数
:return:
"""
result_list = ['not like', 'general like', 'very like']
game_percent = float(input('percentage of time spent playing video games?'))
fly_miles = float(input('frequent flier miles earned per year?'))
how_much_ice_cream = float(input('liters of ice cream consumed per week?'))
dating_data_arr, dating_labels = file2array('resource/datingTestSet2.txt')
norm_arr, ranges, min_vals = auto_norm(dating_data_arr)
in_arr = array([fly_miles, game_percent, how_much_ice_cream])
classifier_result = classify((in_arr - min_vals) / ranges, norm_arr, dating_labels, 3)
print('You will probably like this person: %s' % result_list[classifier_result - 1])
def img2vector(filename):
"""
将32*32的二进制图像矩阵转化为1*1024的向量
:param filename:
:return:
"""
return_vector = zeros((1, 1024))
fr = open(filename)
for i in range(32):
line = fr.readline()
for j in range(32):
return_vector[0, 32*i+j] = int(line[j])
return return_vector
def handwriting_class_test():
handwriting_labels = []
# 训练数据
training_file_list = listdir('resource/digits/trainingDigits')
length = len(training_file_list)
training_arr = zeros((length, 1024))
for i in range(length):
# 获取文件名,含后缀
filename_str = training_file_list[i]
file_str = filename_str.split('.')[0]
# 获取文件中存储二进制图像,表示的数字
class_num_str = int(file_str.split('_')[0])
handwriting_labels.append(class_num_str)
# 将各文件生成的1*1024向量分别存入training_arr
training_arr[i, :] = img2vector('resource/digits/trainingDigits/%s' % filename_str)
# 测试数据
test_file_list = listdir('resource/digits/testDigits')
error_count = 0.0
length = len(test_file_list)
for i in range(length):
filename_str = test_file_list[i]
file_str = filename_str.split('.')[0]
class_num_str = int(file_str.split('_')[0])
# 读取一个测试文件,并生成1*1024的向量,赋值给vector_under_test
vector_under_test = img2vector('resource/digits/testDigits/%s' % filename_str)
classifier_result = classify(vector_under_test, training_arr, handwriting_labels, 3)
print('The classifier came back with: %d,the real answer is: %d'
% (classifier_result, class_num_str))
if classifier_result != class_num_str:
error_count += 1.0
print('The total number of errors is: %d' % error_count)
print('The total error rate is: %f' % (error_count/length))
if __name__ == '__main__':
# show_data_in_chart()
# 通过调整test_ratio的比例,以及k的值,使结果最优
# dating_class_test()
# classify_person()
# test_vector = img2vector('resource/digits/testDigits/0_0.txt')
# print(test_vector[0, 0:32])
handwriting_class_test()
|