└── knn.py


/knn.py:
--------------------------------------------------------------------------------
 1 | # -*- coding: utf-8 -*-
 2 | # 手写数字分类
 3 | from sklearn.model_selection import train_test_split
 4 | from sklearn import preprocessing
 5 | from sklearn.metrics import accuracy_score
 6 | from sklearn.datasets import load_digits
 7 | from sklearn.neighbors import KNeighborsClassifier
 8 | from sklearn.svm import SVC
 9 | from sklearn.naive_bayes import MultinomialNB
10 | from sklearn.tree import DecisionTreeClassifier
11 | import matplotlib.pyplot as plt
12 | 
13 | # 加载数据
14 | digits = load_digits()
15 | data = digits.data
16 | # 数据探索
17 | print(data.shape)
18 | # 查看第一幅图像
19 | print(digits.images[0])
20 | # 第一幅图像代表的数字含义
21 | print(digits.target[0])
22 | # 将第一幅图像显示出来
23 | plt.gray()
24 | plt.imshow(digits.images[0])
25 | plt.show()
26 | 
27 | # 分割数据，将25%的数据作为测试集，其余作为训练集
28 | train_x, test_x, train_y, test_y = train_test_split(data, digits.target, test_size=0.25, random_state=33)
29 | 
30 | # 采用Z-Score规范化
31 | ss = preprocessing.StandardScaler()
32 | train_ss_x = ss.fit_transform(train_x)
33 | test_ss_x = ss.transform(test_x)
34 | 
35 | # 创建KNN分类器
36 | knn = KNeighborsClassifier()
37 | knn.fit(train_ss_x, train_y) 
38 | predict_y = knn.predict(test_ss_x) 
39 | print("KNN准确率: %.4lf" % accuracy_score(test_y, predict_y))
40 | 
41 | # 创建SVM分类器
42 | svm = SVC()
43 | svm.fit(train_ss_x, train_y)
44 | predict_y=svm.predict(test_ss_x)
45 | print('SVM准确率: %0.4lf' % accuracy_score(test_y, predict_y))
46 | 
47 | # 采用Min-Max规范化
48 | mm = preprocessing.MinMaxScaler()
49 | train_mm_x = mm.fit_transform(train_x)
50 | test_mm_x = mm.transform(test_x)
51 | 
52 | # 创建Naive Bayes分类器
53 | mnb = MultinomialNB()
54 | mnb.fit(train_mm_x, train_y) 
55 | predict_y = mnb.predict(test_mm_x) 
56 | print("多项式朴素贝叶斯准确率: %.4lf" % accuracy_score(test_y, predict_y))
57 | 
58 | # 创建CART决策树分类器
59 | dtc = DecisionTreeClassifier()
60 | dtc.fit(train_mm_x, train_y) 
61 | predict_y = dtc.predict(test_mm_x) 
62 | print("CART决策树准确率: %.4lf" % accuracy_score(test_y, predict_y))


--------------------------------------------------------------------------------