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
| import numpy as np
import pandas as pd
"""
load_data_set():
功能：载入数据集
输入：无
返回：
posting_list：数据集；
classes_list：各属性的类别
property_list：各个属性的属性值集合列表
"""
def csv_input(CSV_FILE_PATH):
df = pd.read_csv(CSV_FILE_PATH)
return df.values.tolist()
def load_data_set():
posting_list=csv_input('./西瓜的数据集.csv')
property_list = [
'青绿', '乌黑', '浅白',
'蜷缩', '稍缩', '硬挺',
'浊响', '沉闷', '清脆',
'清晰', '稍糊', '模糊',
'凹陷', '稍陷', '稍凹',
'硬滑', '软粘']
return posting_list, property_list
"""
getNums(posting_list,col,rows,nums):
功能：取值函数，从数据集中取出某一列（密度或者含糖率样本）多行的值，返回一维数组，及其长度
输入：posting_list：数据集；col：所取数据集的列号；rows:所取数据集的开始行号；nums：取的数据行数；
输出：Nums：浮点型数据列表
"""
def getNums(posting_list, col, rows, nums):
Nums = [0] * nums
for n in range(0, nums):
Nums[n] = float(posting_list[rows + n][col])
return Nums
"""
train_naive_bayes(posting_list,property_list):
功能：训练数据，即计算数据集的类条件概率，类先验概率
输入：posting_list：数据集；property_list：各个属性的属性值集合构成的列表
输出：propertyConditionalProbabilityPositive：正样本(好瓜)类条件概率
propertyConditionalProbabilityNegative：负样本类条件概率
"""
def train_naive_bayes(posting_list, property_list):
trainNum = len(posting_list)
pSampleNum = 0
for sample in posting_list:
if sample[-1] == '是':
pSampleNum += 1
prioClass = pSampleNum / trainNum
propertyConditionalProbabilityPositive = []
propertyConditionalProbabilityNegative = []
for propertyy in property_list:
if (propertyy == "硬滑") or (propertyy == "软粘"):
pSampleNumLap = pSampleNum + 2
nSampleNumLap = trainNum - pSampleNum + 2
else:
pSampleNumLap = pSampleNum + 3
nSampleNumLap = trainNum - pSampleNum + 3
posNumPropertyPositive = 1
negNumPropertyPositive = 1
for rows in range(0, len(posting_list)):
if propertyy in posting_list[rows]:
if posting_list[rows][-1] == '是':
posNumPropertyPositive += 1
else:
negNumPropertyPositive += 1
propertyConditionalProbabilityPositive.append(posNumPropertyPositive / pSampleNumLap)
propertyConditionalProbabilityNegative.append(negNumPropertyPositive / nSampleNumLap)
return propertyConditionalProbabilityPositive, propertyConditionalProbabilityNegative
"""
classify_naive_bayes(data,propertyConditionalProbabilityPositive,property_list,propertyConditionalProbabilityNegative):
功能：求正负类别的概率，返回1或者0， 1表示为正样本
输入：data：想要测试的数据，格式见底部说明。propertyConditionalProbabilityPositive正类条件概率；propertyConditionalProbabilityNegative负类条件概率
输出：返回1或者0；其中1代表正样本（好瓜），0代表负样本
"""
def classify_naive_bayes(data, propertyConditionalProbabilityPositive, property_list,
propertyConditionalProbabilityNegative):
probabilityPos = 0
probabilityNeg = 0
for propertyData in data[:-1]:
if propertyData in property_list:
index = property_list.index(propertyData)
probabilityPos += np.log(propertyConditionalProbabilityPositive[index])
probabilityNeg += np.log(propertyConditionalProbabilityNegative[index])
if probabilityPos > probabilityNeg:
return 1
else:
return 0
if __name__ == "__main__":
posting_list, property_list = load_data_set()
propertyConditionalProbabilityPositive, propertyConditionalProbabilityNegative = train_naive_bayes(posting_list,
property_list)
data=[]
data=csv_input('./西瓜的测试集.csv')
accuracy=0
for n in range(0,len(data)):
result = classify_naive_bayes(data[n], propertyConditionalProbabilityPositive, property_list,
propertyConditionalProbabilityNegative)
if result==1 and data[n][-1]=='是' or result==0 and data[n][-1]=='否':
accuracy+=1
print('is it the good melon : {} {}'.format(result,data[n][-1]))
print('准确率为：{}'.format(accuracy/len(data)) )
|