机器学习——CART决策树——泰坦尼克还生还预测

男娘i 2023-09-29 10:33 11阅读 0赞

利用CART分类器进行预测

读取数据

import pandas as pd
    data =pd.read_csv("train.csv")

查看数据

# 显示前五行
    data.head()
    
    # 显示行数和列数
    data.shape
    
    # 显示所有列的数据类型等信息
    data.info()

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16][]

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 1][]

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_19_color_FFFFFF_t_70_g_se_x_16][]

# 显示类别Embarked特征列的所有取值及出现次数
    data.Embarked.value_counts()

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 2][]

### 三、数据处理 ###

1、缺失值处理

在查看数据发现特征Age和Embarked有缺失值

特征Age使用其均值补充

特征Embarked使用众数“S”进行补充

2、特征编码转换

大部分模型只能处理数值型数据

使用将非数值型转换可计算的编码

采用特征编码转换

生成N个二值特征列（取值0或1），每个对应一种取值

使用决策树模型，一般无须对特征进行缩放

# 缺失值处理
    data.Age.fillna(data.Age.median(),inplace=True)
    data.Embarked.fillna('S',inplace=True)
    
    # 特征编码转换
    data.Sex=data.Sex.map({'female':0,'male':1})
    embarked_d=pd.get_dummies(data.Embarked,prefix='Embarked',drop_first=True)
    data=pd.concat([data,embarked_d],axis=1)
    
    # 将处理好的数据放入
    feature_cols=['Pclass','Sex','Age','Embarked_Q','Embarked_S']
    X=data[feature_cols]
    y=data.Survived

###  ###

### 四、训练和选择模型 ###

数据集进行训练

from sklearn.tree import DecisionTreeClassifier
    treeclf = DecisionTreeClassifier(max_depth=3,random_state=1)
    
    treeclf.fit(X,y)

### 五、可视化决策树 ###

import graphviz
    from sklearn import tree
    from graphviz import Digraph
    dot_data = tree.export_graphviz(treeclf,out_file=None,feature_names=feature_cols,class_names='Survived',filled=True,rounded=True,special_characters=True)
    graph=graphviz.Source(dot_data)
    graph.render('xgboost1') #输出pdf文件
    graph

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 3][]

### 六、查看特征的重要性 ###

pd.DataFrame({'feature':feature_cols,'importance':treeclf.feature_importances_})

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 4][]

### 七、模型评估 ###

from sklearn.model_selection import train_test_split
    X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.7, test_size=0.3, random_state=4)
    
    
    
    from sklearn.model_selection import GridSearchCV
    parameters = {'max_depth':[1,3,5,10,15,20,30]}
    tree_clf=GridSearchCV(DecisionTreeClassifier(),param_grid=parameters,scoring='accuracy')
    tree_clf.fit(X_train,y_train)

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 5][]

print(tree_clf.best_params_)
    print(tree_clf.best_score_)

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 6][]

y_pred=tree_clf.predict(X_test)
    
    from sklearn.metrics import accuracy_score
    from sklearn.metrics import classification_report
    print(accuracy_score(y_test,y_pred))
    print(classification_report(y_test,y_pred))

![watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 7][]

### 完整代码： ###

import pandas as pd
    data =pd.read_csv("train.csv")
    
    data.Age.fillna(data.Age.median(),inplace=True)
    data.Embarked.fillna('S',inplace=True)
    
    data.Sex=data.Sex.map({'female':0,'male':1})
    embarked_d=pd.get_dummies(data.Embarked,prefix='Embarked',drop_first=True)
    data=pd.concat([data,embarked_d],axis=1)
    
    feature_cols=['Pclass','Sex','Age','Embarked_Q','Embarked_S']
    X=data[feature_cols]
    y=data.Survived
    
    from sklearn.tree import DecisionTreeClassifier
    treeclf = DecisionTreeClassifier(max_depth=3,random_state=1)
    treeclf.fit(X,y)
    
    import graphviz
    from sklearn import tree
    from graphviz import Digraph
    dot_data = tree.export_graphviz(treeclf,out_file=None,feature_names=feature_cols,class_names='Survived',filled=True,rounded=True,special_characters=True)
    graph=graphviz.Source(dot_data)
    graph
    
    
    
    pd.DataFrame({'feature':feature_cols,'importance':treeclf.feature_importances_})
    from sklearn.model_selection import train_test_split
    X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=0.7, test_size=0.3, random_state=4)
    
    from sklearn.model_selection import GridSearchCV
    from sklearn.tree import DecisionTreeClassifier
    
    parameters = {'max_depth':[1,3,5,10,15,20,30]}
    tree_clf=GridSearchCV(DecisionTreeClassifier(),param_grid=parameters,scoring='accuracy')
    tree_clf.fit(X_train,y_train)
    
    
    print(tree_clf.best_params_)
    print(tree_clf.best_score_)
    
    y_pred=tree_clf.predict(X_test)
    from sklearn.metrics import accuracy_score
    print(accuracy_score(y_test,y_pred))
    print(classification_report(y_test,y_pred))

[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16]: https://img-blog.csdnimg.cn/79311c0abebf4610b30f9920672374b8.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 1]: https://img-blog.csdnimg.cn/e63990ab757b4bb98aa6ec7563c7af06.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_19_color_FFFFFF_t_70_g_se_x_16]: https://img-blog.csdnimg.cn/c61c56634cfa4b24ae0207350e3f7a27.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_19,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 2]: https://img-blog.csdnimg.cn/5eda03f18c0343cb9130f8e1d5ede7db.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 3]: https://img-blog.csdnimg.cn/b9e55e80980946c8936988acb08b7577.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 4]: https://img-blog.csdnimg.cn/bb25876cd2f248f5b1b551c25e067659.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 5]: https://img-blog.csdnimg.cn/1d91e489cb3a4d0eb8d28c7e6f91eb54.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 6]: https://img-blog.csdnimg.cn/cea18b3c4c2246f5a431c63bc3eb6104.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16
[watermark_type_d3F5LXplbmhlaQ_shadow_50_text_Q1NETiBA54y_X-WQjOWtpg_size_20_color_FFFFFF_t_70_g_se_x_16 7]: https://img-blog.csdnimg.cn/2d36807980c541598955727d693780c9.png?x-oss-process=image/watermark,type_d3F5LXplbmhlaQ,shadow_50,text_Q1NETiBA54y_X-WQjOWtpg==,size_20,color_FFFFFF,t_70,g_se,x_16