经纬度聚类

1. 经纬度数据

xy = np.array([[29.060703, 113.348653], [22.5445697, 114.0545346], 
              [43.8882168, 125.3182578], [30.5951051, 114.2999353], 
              [26.0774954, 119.2918215], [23.0444712, 113.7465512], 
              [42.0151733, 121.6452868], [32.0609736, 118.7916458], 
              [22.5445697, 114.0545346], [39.1235635, 117.1980785], 
              [36.6507007, 117.1140042], [31.2252985, 121.4890497], 
              [20.0423626, 110.3409936], [22.5213807, 113.3656141], 
              [29.3595966, 115.9428552], [26.5814112, 106.7052544], 
              [23.1301964, 113.2592945], [41.494444, 121.615278], 
              [23.0444712, 113.7465512], [39.906217, 116.3912757], 
              [22.5213807, 113.3656141], [9.6153904, 112.970189230798], 
              [39.1235635, 117.1980785], [30.466238, 111.686209455868], 
              [45.7656666, 126.6160584], [32.5664035, 128.899626584395], 
              [43.8882168, 125.3182578], [23.0515551, 112.4728451], 
              [31.32640095, 120.990580800174], [27.9958689, 120.6915792], 
              [39.906217, 116.3912757], [22.5445697, 114.0545346], 
              [35.218039, 111.053715], [39.906217, 116.3912757], 
              [31.9827896, 120.8904588], [31.458932, 104.750108], 
              [34.7591877, 113.6524076], [27.9958689, 120.6915792], 
              [39.906217, 116.3912757], [39.1235635, 117.1980785], 
              [32.0609736, 118.7916458], [31.8129864, 119.9692045], 
              [31.8782976, 120.5514595], [39.1235635, 117.1980785], 
              [29.5585712, 106.5492822], [46.8087794, 130.3734556], 
              [31.2252985, 121.4890497], [39.1235635, 117.1980785], 
              [34.3468421, 108.9365887], [22.5445697, 114.0545346], 
              [31.8129864, 119.9692045], [39.906217, 116.3912757], 
              [33.387124, 104.9313723], [31.2252985, 121.4890497], 
              [25.4342199, 119.0074625], [43.8254059, 87.614569], 
              [39.906217, 116.3912757], [39.906217, 116.3912757], 
              [24.4586961, 118.0745391], [40.01363995, 119.35495441411], 
              [34.7591877, 113.6524076], [30.5951051, 114.2999353], 
              [30.65093955, 120.783165577059], [39.906217, 116.3912757], 
              [30.5951051, 114.2999353], [29.5585712, 106.5492822], 
              [23.1301964, 113.2592945], [39.1235635, 117.1980785], 
              [28.6581723, 121.4163876], [40.6549157, 109.8344586], 
              [39.906217, 116.3912757], [22.8211654, 108.3159756], 
              [32.0609736, 118.7916458], [23.1301964, 113.2592945], 
              [38.04, 114.47], [30.7031196, 111.2750181], 
              [34.0472604, 113.804958141593], [26.0774954, 119.2918215], 
              [34.3468421, 108.9365887], [39.906217, 116.3912757], 
              [23.0247687, 113.1146335], [27.13543065, 105.565762465037], 
              [32.0609736, 118.7916458], [43.6392011, 122.2497308], 
              [29.8622194, 121.6203873], [30.2134399, 115.0670135], 
              [38.21211485, 116.838327521023], [21.9745252, 113.931350084859], 
              [29.7275361, 115.9913134], [31.2252985, 121.4890497], 
              [28.1979483, 112.97133], [26.6655699, 119.5224895], 
              [29.75929015, 120.51582161865], [29.5585712, 106.5492822], 
              [36.4441808, 115.9642007]]
    )

2. 坐标聚类生成子图

for k in range(2,6):    #选择连续不同的 k 值

    matplotlib.rcParams['font.sans-serif']=['SimHei']
    
    if k == 2:
        ax = fig.add_subplot(221)
        ax.set_title('聚类图(2个簇点)')
    if k == 3:
        ax = fig.add_subplot(222)
        ax.set_title('聚类图(3个簇点)')
    if k == 4:
        ax = fig.add_subplot(223)
        ax.set_title('聚类图(4个簇点)')
    if k == 5:
        ax = fig.add_subplot(224)
        ax.set_title('聚类图(5个簇点)')
        
    clf = KMeans(n_clusters=k) #n_clusters用于指定聚类的簇数
    s = clf.fit(xy) #加载数据集合
    numSamples=len(xy)
    centroids = clf.labels_
   
    mark = ['or', 'ob', 'og', 'ok', '^r', '+r', 'sr', 'dr', '<r', 'pr']
    #画出所有样例点 属于同一分类的绘制同样的颜色
    for i in range(numSamples):
        ax.plot(xy[i][1], xy[i][0], mark[clf.labels_[i]]) #mark[markIndex])
        
    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):
        ax.plot(centroids[i][1], centroids[i][0], mark[i], markersize = 12)
        
plt.subplots_adjust(wspace =0.3, hspace =0.5)#调整子图间距
plt.savefig('聚类图.png', dpi=200)  

plt.show()   

3. 结果展示