Python数据分析_第13课：分类算法_笔记

淩亂°似流年 2021-10-29 06:58 489阅读 0赞

### 文章目录 ###

*  分类算法案例：预测饭店销量
 *   *  导入数据
     *  拆分训练数据与测试数据
     *  训练KNN分类器
     *   *  测试结果
         *  准确率
     *  训练贝叶斯分类器
     *   *  测试结果
         *  准确率
     *  决策树
     *   *  测试结果
         *  准确率
     *  SVM
     *   *  测试结果
         *  准确率

GitHub: [https://github.com/RealEmperor/Python-for-Data-Analysis][https_github.com_RealEmperor_Python-for-Data-Analysis]

分类算法有很多，这里通过一个案例介绍一下几个常用算法的应用，包括：K最近邻(kNN，k-NearestNeighbor)分类算法、贝叶斯分类算法、决策树、支持向量机（Support Vector Machine, SVM）。

# 分类算法案例：预测饭店销量 #

## 导入数据 ##

import numpy as np
    import pandas as pd
    from sklearn.metrics import classification_report
    from sklearn.metrics import precision_recall_curve
    from sklearn.model_selection import train_test_split
    from sklearn.naive_bayes import BernoulliNB
    from sklearn.neighbors import KNeighborsClassifier
    
    inputfile = 'data/sales_data.xls'
    data = pd.read_excel(inputfile, index_col=u'序号')  # 导入数据
    data

<table> 
 <thead> 
  <tr> 
   <th></th> 
   <th>天气</th> 
   <th>是否周末</th> 
   <th>是否有促销</th> 
   <th>销量</th> 
  </tr> 
  <tr> 
   <th>序号</th> 
   <th></th> 
   <th></th> 
   <th></th> 
   <th></th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <th>1</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>2</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>3</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>4</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>5</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>6</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>7</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>否</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>8</th> 
   <td>好</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>9</th> 
   <td>好</td> 
   <td>是</td> 
   <td>否</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>10</th> 
   <td>好</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>11</th> 
   <td>好</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>12</th> 
   <td>好</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>13</th> 
   <td>好</td> 
   <td>是</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>14</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>15</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>16</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>17</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>18</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>19</th> 
   <td>好</td> 
   <td>否</td> 
   <td>否</td> 
   <td>高</td> 
  </tr> 
  <tr> 
   <th>20</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>21</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>22</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>23</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>24</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>25</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>26</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>27</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>28</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>29</th> 
   <td>坏</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>30</th> 
   <td>好</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>31</th> 
   <td>坏</td> 
   <td>是</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>32</th> 
   <td>好</td> 
   <td>否</td> 
   <td>是</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>33</th> 
   <td>好</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
  <tr> 
   <th>34</th> 
   <td>好</td> 
   <td>否</td> 
   <td>否</td> 
   <td>低</td> 
  </tr> 
 </tbody> 
</table>

# 数据是类别标签，要将它转换为数据
    # 用1来表示“好”、“是”、“高”这三个属性，用-1来表示“坏”、“否”、“低”
    data[data == u'好'] = 1
    data[data == u'是'] = 1
    data[data == u'高'] = 1
    data[data != 1] = -1
    x = data.iloc[:, :3].as_matrix().astype(int)
    print(x)
    y = data.iloc[:, 3].as_matrix().astype(int)
    print(y)

[[-1  1  1]
     [-1  1  1]
     [-1  1  1]
     [-1 -1  1]
     [-1  1  1]
     [-1 -1  1]
     [-1  1 -1]
     [ 1  1  1]
     [ 1  1 -1]
     [ 1  1  1]
     [ 1  1  1]
     [ 1  1  1]
     [ 1  1  1]
     [-1  1  1]
     [ 1 -1  1]
     [ 1 -1  1]
     [ 1 -1  1]
     [ 1 -1  1]
     [ 1 -1 -1]
     [-1 -1 -1]
     [-1 -1  1]
     [-1 -1  1]
     [-1 -1  1]
     [-1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1 -1  1]
     [-1 -1 -1]
     [-1 -1 -1]
     [ 1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1 -1 -1]
     [ 1 -1 -1]]
    [ 1  1  1  1  1  1  1  1  1  1  1  1  1 -1  1  1  1  1  1 -1 -1 -1 -1 -1 -1
     -1 -1 -1 -1 -1 -1 -1 -1 -1]

## 拆分训练数据与测试数据 ##

# 拆分训练数据与测试数据
    x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.2)

## 训练KNN分类器 ##

clf = KNeighborsClassifier(algorithm='kd_tree')
    clf.fit(x_train, y_train)

KNeighborsClassifier(algorithm='kd_tree', leaf_size=30, metric='minkowski',
               metric_params=None, n_jobs=1, n_neighbors=5, p=2,
               weights='uniform')

### 测试结果 ###

# 测试结果
    answer = clf.predict(x_test)
    print(x_test)
    print(answer)
    print(y_test)
    print(np.mean(answer == y_test))

[[ 1 -1  1]
     [-1  1  1]
     [-1 -1  1]
     [ 1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1  1  1]]
    [-1  1 -1 -1 -1 -1  1]
    [ 1  1  1  1 -1  1  1]
    0.428571428571

### 准确率 ###

# 准确率
    precision, recall, thresholds = precision_recall_curve(y_train, clf.predict(x_train))
    print(classification_report(y_test, answer, target_names=['高', '低']))

precision    recall  f1-score   support
    
              高       0.20      1.00      0.33         1
              低       1.00      0.33      0.50         6
    
    avg / total       0.89      0.43      0.48         7

## 训练贝叶斯分类器 ##

# 训练贝叶斯分类器
    clf = BernoulliNB()
    clf.fit(x_train, y_train)

BernoulliNB(alpha=1.0, binarize=0.0, class_prior=None, fit_prior=True)

### 测试结果 ###

# 测试结果
    answer = clf.predict(x_test)
    print(x_test)
    print(answer)
    print(y_test)
    print(np.mean(answer == y_test))

[[ 1 -1  1]
     [-1  1  1]
     [-1 -1  1]
     [ 1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1  1  1]]
    [-1  1 -1 -1 -1 -1  1]
    [ 1  1  1  1 -1  1  1]
    0.428571428571

### 准确率 ###

print(classification_report(y_test, answer, target_names=['低', '高']))

precision    recall  f1-score   support
    
              低       0.20      1.00      0.33         1
              高       1.00      0.33      0.50         6
    
    avg / total       0.89      0.43      0.48         7

## 决策树 ##

from sklearn.tree import DecisionTreeClassifier as DTC
    
    dtc = DTC(criterion='entropy')  # 建立决策树模型，基于信息熵
    dtc.fit(x_train, y_train)  # 训练模型
    
    # 导入相关函数，可视化决策树。
    # 导出的结果是一个dot文件，需要安装Graphviz才能将它转换为pdf或png等格式。
    
    # https://graphviz.gitlab.io/_pages/Download/Download_windows.html
    # 安装Graphviz https://graphviz.gitlab.io/_pages/Download/windows/graphviz-2.38.msi
    
    from sklearn.tree import export_graphviz
    
    with open("data/tree.dot", 'w') as f:
        f = export_graphviz(dtc, out_file=f)

### 测试结果 ###

# 测试结果
    answer = dtc.predict(x_test)
    print(x_test)
    print(answer)
    print(y_test)
    print(np.mean(answer == y_test))

[[ 1 -1  1]
     [-1  1  1]
     [-1 -1  1]
     [ 1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1  1  1]]
    [-1  1 -1 -1 -1 -1  1]
    [ 1  1  1  1 -1  1  1]
    0.428571428571

### 准确率 ###

print(classification_report(y_test, answer, target_names=['低', '高']))

## SVM ##

from sklearn.svm import SVC
    
    clf = SVC()
    clf.fit(x_train, y_train)

SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0,
      decision_function_shape=None, degree=3, gamma='auto', kernel='rbf',
      max_iter=-1, probability=False, random_state=None, shrinking=True,
      tol=0.001, verbose=False)

### 测试结果 ###

# 测试结果
    answer = clf.predict(x_test)
    print(x_test)
    print(answer)
    print(y_test)
    print(np.mean(answer == y_test))

[[ 1 -1  1]
     [-1  1  1]
     [-1 -1  1]
     [ 1 -1 -1]
     [-1  1 -1]
     [ 1 -1  1]
     [ 1  1  1]]
    [-1  1 -1 -1  1 -1  1]
    [ 1  1  1  1 -1  1  1]
    0.285714285714

### 准确率 ###

print(classification_report(y_test, answer, target_names=['低', '高']))

precision    recall  f1-score   support
    
              低       0.00      0.00      0.00         1
              高       0.67      0.33      0.44         6
    
    avg / total       0.57      0.29      0.38         7

参考资料：炼数成金Python数据分析课程

[https_github.com_RealEmperor_Python-for-Data-Analysis]: https://github.com/RealEmperor/Python-for-Data-Analysis