import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import DBSCAN

dataS = np.genfromtxt('testSet.txt')
plt.scatter(dataS[:,0],dataS[:,1])
plt.show()

#建立模型，其中eps代表距离阀值，min_samples代表核心对象在eps领域的样本数阀值
#dbscan = DBSCAN(eps=1,min_samples=4)
dbscan = DBSCAN(eps=1.5,min_samples=5)
dbscan.fit(dataS)


result = dbscan.fit_predict(dataS)
print(result)

#画图
plt.rcParams['font.sans-serif']=['SimHei']#用来正常显中文标签，SimHei代表黑体
plt.rcParams['axes.unicode_minus']=False#用来正常显示负号，minus代表负号
plt.scatter(dataS[result==-1,0],dataS[result==-1,1],s=150,c='purple',marker='8',label='噪声点')


plt.scatter(dataS[result==0,0],dataS[result==0,1],s=150,c='orange',marker='o',label='cluster=1')
plt.scatter(dataS[result==1,0],dataS[result==1,1],s=150,c='green',marker='s',label='cluster=2')
plt.scatter(dataS[result==2,0],dataS[result==2,1],s=150,c='blue', marker='^',label='cluster=3')
plt.scatter(dataS[result==3,0],dataS[result==3,1],s=150,c='red',marker='*',label='cluster=4')
plt.scatter(dataS[result==4,0],dataS[result==4,1],s=150,c='black',marker='p',label='cluster=5')

#p代表五边形



plt.legend(loc='lower left',bbox_to_anchor=(-0.4,0.5,0.1,0.1))
plt.show()

x1,y1 = datasets.make_circles(n_samples=2000,factor=0.5,noise=0.05)#make_circles生成环
x2,y2 = datasets.make_blobs(n_samples=1000,centers=[[1.2,1.2]],cluster_std=[[.1]])
#factor代表的是内外圈之间的比例因子，nosie代表噪音
#make_blobs是聚类数据生成器，cluster_std是聚类的标准查，课调节样本点的分散程度
x=np.concatenate((x1,x2))
plt.scatter(x[:,0],x[:,1],marker='o')
plt.show()

#定义模型
dbscan = DBSCAN(eps=0.2,min_samples=50)
#训练模型及对样本点进行聚类

cluster=dbscan.fit_predict(x)
print(cluster)
plt.scatter(x[:,0],x[:,1],c=cluster)
plt.show()