機器學習(一): python三種特徵選擇方法
阿新 • • 發佈:2018-12-31
特徵選擇的三種方法介紹:
過濾型:
選擇與目標變數相關性較強的特徵。缺點:忽略了特徵之間的關聯性。
包裹型:
基於線性模型相關係數以及模型結果AUC逐步剔除特徵。如果剔除相關係數絕對值較小特徵後,AUC無大的變化,或降低,則可剔除
嵌入型:
利用模型提取特徵,一般基於線性模型與正則化(正則化取L1),取權重非0的特徵。(特徵緯度特別高,特別稀疏,用svd,pca算不動)
python 實現
"""1.過濾型"""
from sklearn.datasets import load_iris
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2
iris=load_iris()
X,y=iris.data,iris.target
print X.shape
X_new=SelectKBest(chi2,k=2).fit_transform(X,y)
print X_new.shape
"""輸出:
(150L, 4L)
(150L, 2L)"""
"""2.包裹型"""
from sklearn.feature_selection import RFE
from sklearn.linear_model import LinearRegression
from sklearn.datasets import load_boston
boston=load_boston()
X=boston["data"]
Y=boston["target"]
names=boston["feature_names"]
lr=LinearRegression()
rfe=RFE(lr,n_features_to_select=1)#選擇剔除1個
rfe.fit(X,Y)
print "features sorted by their rank:"
print sorted(zip(map(lambda x:round(x,4), rfe.ranking_),names))
"""輸出:按剔除後AUC排名給出
features sorted by their rank:
[(1.0, 'NOX'), (2.0, 'RM'), (3.0, 'CHAS'), (4.0, 'PTRATIO'), (5.0, 'DIS'), (6.0, 'LSTAT'), (7.0, 'RAD'), (8.0, 'CRIM'), (9.0, 'INDUS'), (10.0, 'ZN'), (11.0, 'TAX')
, (12.0, 'B'), (13.0, 'AGE')]"""
"""3.嵌入型 ,老的版本沒有SelectFromModel"""
from sklearn.svm import LinearSVC
from sklearn.datasets import load_iris
from sklearn.feature_selection import SelectFromModel
iris=load_iris()
X,y=iris.data,iris.target
print X.shape
lsvc=LinearSVC(C=0.01,penalty='l1',dual=False).fit(X,y)
model=SelectFromModel(lsvc,prefit=True)
X_new=model.transform(X)
print X_new.shape
"""輸出:
(150,4)
(150,3)
"""