Python机器学习与深度学习之三：数据特征选择

一、数据特征选择目的

1、降低数据的拟合度

2、提高算法精度

3、减少训练时间

二、数据特征方法

1、单变量特征（卡方检验）

import pandas as pd
import numpy as np
from sklearn import datasets
from sklearn.feature_selection import SelectKBest
from sklearn.feature_selection import chi2

iris  = datasets.load_iris()

names = ['separ-length','separ-width','petal-length','petal-width','class']
data  = pd.read_csv(r'iris.csv',names = names)

array = data.values
X     = array[:,0:4]
Y     = array[:,4]

test  = SelectKBest(score_func = chi2, k = 4)
fit   = test.fit(X,Y)

np.set_printoptions(precision = 3)

print(fit.scores_)
features = fit.transform(X)

print(features)

运行结果：

2、递归特征消除

下面代码以支持向量机算法模型，通过 递归特征消除来选定对预测结果影响最大的三个数据特征

import pandas as pd
import numpy as np
from sklearn import datasets
from sklearn.feature_selection import RFE
from sklearn.svm import SVC

iris  = datasets.load_iris()

names = ['separ-length','separ-width','petal-length','petal-width','class']
data  = pd.read_csv(r'iris.csv',names = names)

array = data.values
X     = array[:,0:4]
Y     = array[:,4]

model = SVC(kernel="linear")
rfe   = RFE(model,3)
fit   = rfe.fit(X,Y)

print("特征个数:")
print(fit.n_features_)
print("被选定的特征：")

print(fit.support_)
print("特征排名：")
print(fit.ranking_)

运行结果：