调用sklearn.MeanShift聚类算法的测试

调用sklearn.MeanShift聚类算法的测试

#导入相关的模块
from sklearn.cluster import KMeans
from sklearn.externals import joblib
import numpy, time
import matplotlib.pyplot as plt
from sklearn.cluster import MeanShift, estimate_bandwidth
import numpy as np

if __name__=='__main__':
    #加载数据
    print ('step1:load data')
    
    dataset=[]
    fileIn = open(r'E:\项目数据\test_mean_shift.txt')
    for line in fileIn.readlines():
        lineArr=line.strip().split(' ')
        #print(lineArr[0])
        dataset.append([float(lineArr[0]),float(lineArr[1])])
    
    numSamples=len(dataset)  #得到样本数据的数量
    print(numSamples)
    X=np.array(dataset) #列表类型转换为数组类型
    #print(X)
    bandwidth=estimate_bandwidth(X, quantile=0.2, n_samples=500)  #定义以某个点为核心时搜索半径
    #下一步是训练数据，并且将变量clf实例化
    clf=MeanShift(bandwidth=bandwidth, bin_seeding=True, cluster_all=True).fit(X)
    
    centroids=clf.labels_
    print(centroids, type(centroids)) #显示每个点聚类归属

	#计算自动生成的K,将聚类数量小于3的排除
    arr_flag=[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0]
    for i in clf.labels_:
        arr_flag[i]=arr_flag[i]+1
    print(arr_flag)
    k=0
    for i in arr_flag:
        if (i>3):
            k+=1
    print(k)
    
    #下一步是构建颜色和样式的列表，for循环中根据每个元素的index选取对应的元素
    mark=['or','ob','og','ok','^r','+r','sr','dr','<r','pr']
    #print(mark[clf.labels_[1]])
    #画出所有样例点，属于同一分类的绘制同样颜色
    for i in range(numSamples):
        plt.plot(dataset[i][0],dataset[i][1],mark[clf.labels_[i]]) #第一个是每个点横坐标，第二个纵坐标，第三个是点的样式
    #下一步重新构建mark列表来定义质心颜色和样式
    mark = ['Dr', 'Db', 'Dg', 'Dk', '^b', '+b', 'sb', 'db', '<b', 'pb']
    # 画出质点，用特殊图型
    centroids =  clf.cluster_centers_
    for i in range(k):
        plt.plot(centroids[i][0], centroids[i][1], mark[i], markersize = 12)
    print (centroids) #显示中心点坐标
    plt.show()

结果展示：