机器学习算法导论---Ksvm--变色鸢尾花预测问题的核方法

import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split
import matplotlib.pyplot as plt
from machine_learning.lib.kernel_svm import KernelSVM

def rbf_kernel(x1,x2):
    sigma=1.0
    return np.exp(-np.linalg.norm(x1-x2,2)**2/sigma)

iris=datasets.load_iris()
X=iris["data"][:,(2,3)]
y=2*(iris["target"]==1).astype(np.int).reshape(-1,1)-1
X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.4,random_state=5)

model=KernelSVM(rbf_kernel)
model.fit(X_train,y_train)

x0s=np.linspace(1,7,100)
x1s=np.linspace(0,3,100)
x0,x1=np.meshgrid(x0s,x1s)
W=np.c_[x0.ravel(),x1.ravel()]
u=model.predict(W).reshape(x0.shape)
plt.plot(X_train[:,0][y_train[:,0]==1],X_train[:,1][y_train[:,0]==1],"bs")
plt.plot(X_train[:,0][y_train[:,0]==-1],X_train[:,1][y_train[:,0]==-1],"yo")
plt.contourf(x0,x1,u,alpha=0.2)
plt.show()