标签：集成 13 clf train 随机 test import voting sklearn

 

13-1什么是集成学习

 

 

 

 

 

 

 

 

 

Notbook 示例

 

 Notbook 源码

 

 1 集成学习
 2 [1]
 3 import numpy as np
 4 import matplotlib.pyplot as plt
 5 [2]
 6 from sklearn import datasets
 7 X, y = datasets.make_moons(n_samples=500,noise=0.3, random_state=42)
 8 [3]
 9 plt.scatter(X[y==0,0], X[y==0,1])
10 plt.scatter(X[y==1,0], X[y==1,1])
11 <matplotlib.collections.PathCollection at 0x2337fd4a160>
12 
13 [4]
14 from sklearn.model_selection import train_test_split
15 
16 X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
17 [5]
18 from sklearn.linear_model import LogisticRegression
19 
20 log_clf = LogisticRegression()
21 log_clf.fit(X_train, y_train)
22 log_clf.score(X_test, y_test)
23 0.864
24 [6]
25 from sklearn.svm import SVC
26 
27 svm_clf = SVC()
28 svm_clf.fit(X_train, y_train)
29 svm_clf.score(X_test, y_test)
30 0.896
31 [7]
32 from sklearn.tree import DecisionTreeClassifier
33 
34 dt_clf = DecisionTreeClassifier()
35 dt_clf.fit(X_train, y_train)
36 dt_clf.score(X_test, y_test)
37 0.84
38 [8]
39 y_predict1 = log_clf.predict(X_test)
40 y_predict2 = svm_clf.predict(X_test)
41 y_predict3 = dt_clf.predict(X_test)
42 [9]
43 y_predict = np.array( (y_predict1 + y_predict2 + y_predict3) >= 2, dtype='int')
44 [10]
45 y_predict[:10]
46 array([1, 0, 0, 1, 1, 1, 0, 0, 0, 0])
47 [11]
48 from sklearn.metrics import accuracy_score
49 
50 accuracy_score(y_test, y_predict)
51 0.904
52 Voting Classifier
53 [12]
54 from sklearn.ensemble import VotingClassifier
55 
56 voting_clf = VotingClassifier(estimators=[
57     ('log_clf', LogisticRegression()),
58     ('svm_clf',SVC()),
59     ('dt_clf', DecisionTreeClassifier())
60 ],voting='hard')
61 [13]
62 voting_clf.fit(X_train, y_train)
63 VotingClassifier(estimators=[('log_clf', LogisticRegression()),
64                              ('svm_clf', SVC()),
65                              ('dt_clf', DecisionTreeClassifier())])
66 [14]
67 voting_clf.score(X_test,y_test)
68 0.912

 

13-2 SoftVoting Classifier

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Notbook 示例

 

 

Notbook 源码

 1 Soft Voting
 2 [1]
 3 import numpy as np
 4 import matplotlib.pyplot as plt
 5 [2]
 6 from sklearn import datasets
 7 X, y = datasets.make_moons(n_samples=500,noise=0.3, random_state=42)
 8 [3]
 9 plt.scatter(X[y==0,0], X[y==0,1])
10 plt.scatter(X[y==1,0], X[y==1,1])
11 <matplotlib.collections.PathCollection at 0x1dbd04d9190>
12 
13 [4]
14 from sklearn.model_selection import train_test_split
15 
16 X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
17 Hard Voting Classifier
18 [5]
19 from sklearn.linear_model import LogisticRegression
20 from sklearn.svm import SVC
21 from sklearn.tree import DecisionTreeClassifier
22 from sklearn.ensemble import VotingClassifier
23 
24 voting_clf = VotingClassifier(estimators=[
25     ('log_clf', LogisticRegression()),
26     ('svm_clf',SVC()),
27     ('dt_clf', DecisionTreeClassifier())
28 ],voting='hard')
29 [6]
30 voting_clf.fit(X_train, y_train)
31 voting_clf.score(X_test,y_test)
32 0.912
33 Soft Voting Classifier
34 [7]
35 voting_clf2 = VotingClassifier(estimators=[
36     ('log_clf', LogisticRegression()),
37     ('svm_clf',SVC(probability=True)),
38     ('dt_clf', DecisionTreeClassifier())
39 ],voting='soft')
40 [8]
41 voting_clf2.fit(X_train, y_train)
42 voting_clf2.score(X_test,y_test)
43 0.92

 

 

 

标签：集成,13,clf,train,随机,test,import,voting,sklearn
From： https://www.cnblogs.com/Cai-Gbro/p/16855753.html