import matplotlib.pyplot as plt
from sklearn import datasets

X,z = datasets.make_circles()
x,y = X.T
plt.axes(aspect=1)
plt.scatter(x,y,c=z,cmap='spring')
plt.show()

plt.figure(figsize=[6,7])
for i,n in enumerate([0.1,0.2,0.3]):
    for j,f in enumerate([0.1,0.4,0.7]):
        X,z = datasets.make_circles(n_samples=150,noise=n,factor=f,random_state=111)
        x,y = X.T
        plt.subplot2grid((3,3),(i,j),xlim=[x.min(),x.max()],ylim=[y.min(),y.max()],xticks=[],yticks=[],aspect=1)
        plt.scatter(x,y,s=20,c=z,cmap='summer',edgecolor='k')
        plt.title('noise=%.1f,factor=%.1f'%(n,f),size=10)
plt.tight_layout()
plt.show()

import numpy as np
from sklearn.linear_model import LogisticRegression as Lori
from sklearn.neighbors import KNeighborsClassifier as Knn
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier as Rafo

X,z = datasets.make_circles(n_samples=200,noise=0.15,factor=0.25)
x,y = X.T
model = [Lori(),Knn(),SVC(),Rafo()]
chue = [u'การถดถอยโลจิสติก',u'เพื่อนบ้านใกล้สุด k ตัว',u'SVM',u'ป่าสุ่ม']
mx,my = np.meshgrid(np.linspace(x.min(),x.max(),200),np.linspace(y.min(),y.max(),200))
mX = np.array([mx.ravel(),my.ravel()]).T
plt.figure(figsize=[7,7])
for i,m in enumerate(model):
    m.fit(X,z)
    mz = m.predict(mX).reshape(200,200)
    plt.subplot(221+i,xlim=[x.min(),x.max()],ylim=[y.min(),y.max()],aspect=1)
    plt.scatter(x,y,c=z,edgecolor='k',cmap='autumn')
    plt.contourf(mx,my,mz,alpha=0.1,cmap='autumn')
    plt.title(chue[i],family='Tahoma')
plt.show()