linearSVC，SC的linear，poly,rbf,sigmoid等效换用具体举例

＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃
首先使用linearSVC
$K(v_1,v_2)=<v_1,v_2>$
Polynomial_no_kernel.py

# -*- coding: utf-8 -*-
import sys
reload(sys)
sys.setdefaultencoding('utf-8')
# @Author: appleyuchi
# @Date:   2018-12-15 20:35:43
# @Last Modified by:   appleyuchi
# @Last Modified time: 2018-12-15 22:20:19
import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets
from plot import *
X, y = datasets.make_moons(noise=0.15, random_state=666)
# plt.scatter(X[y==0, 0], X[y==0, 1])
# plt.scatter(X[y==1, 0], X[y==1, 1])
# plt.show()
print"X=",X
print len(X)
print"y=",y
print len(y)


from sklearn.preprocessing import PolynomialFeatures, StandardScaler
from sklearn.svm import LinearSVC
from sklearn.svm import SVC
from sklearn.pipeline import Pipeline
import sklearn.svm
# print dir(sklearn.svm)

def PolynomialSVC(degree, C=3):
    return Pipeline([
        ('poly', PolynomialFeatures(degree=degree)),
        ('std_scaler', StandardScaler()),
        # ('linearSVC', LinearSVC(C=C))#注意这两句都行
        ('kernelSVC', SVC(kernel='linear', degree=degree, C=C))#注意这两句都行
    ])

poly_svc = PolynomialSVC(degree=5)
poly_svc.fit(X, y)

plot_decision_boundary(poly_svc, axis=[-1.5, 2.5, -1.0, 1.5])
plt.scatter(X[y==0, 0], X[y==0, 1])
plt.scatter(X[y==1, 0], X[y==1, 1])
plt.show()

＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃
接下来使用SVC＋linear核
$K(v_1,v_2)=<v_1,v_2>$
将上面的

('linearSVC', LinearSVC(C=C))

换成

('kernelSVC', SVC(kernel='linear', degree=degree, C=C))

效果是一样的:

＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃
接下来使用SVC+poly核
$K(v_1,v_2)=(\gamma <v_1,v_2>+c{)}^n$
将Polynomial_no_kernel.py文件中的内容改为：

def PolynomialSVC(degree, C=1):
    return Pipeline([
        ('poly', PolynomialFeatures(degree=degree)),
        ('std_scaler', StandardScaler()),
        # ('linearSVC', LinearSVC(C=C))#注意这两句都行
        ('kernelSVC', SVC(kernel='poly', degree=degree, C=C,class_weight={0:40,1:70})
        )#注意这两句都行
    ])

poly_svc = PolynomialSVC(degree=2)
poly_svc.fit(X, y)

效果如下：

＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃
接下来使用SVC+rbf核：
$K(v_1,v_2)=\exp (-\gamma ||v_1-v_2|{|}^2)$
将

('kernelSVC', SVC(kernel='poly', degree=degree, C=C,class_weight={0:40,1:70})

改为：

('kernelSVC', SVC(kernel='rbf', degree=degree, C=C,class_weight={0:40,1:70})

效果如下:

#＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃＃
最后使用SVC＋sigmoid核
修改其中两句代码为：

        ('kernelSVC', SVC(kernel='sigmoid', degree=degree, C=C,class_weight={0:80,1:80})
        
        poly_svc = PolynomialSVC(degree=10)

结果如下：

完整代码如下：
https://github.com/appleyuchi/scikit_kernel_use

参考连接：
https://www.cnblogs.com/volcao/p/9464991.html
https://blog.csdn.net/qq_41577045/article/details/79859902