聚类_Kmeans优化

代价函数最小

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris
def distance(vex1,vex2):
    return np.sqrt(np.sum(np.power(vex1-vex2,2)))
def kMeans_way(S,k,distMeas=distance):
    m=np.shape(S)[0]
    sampleTag = np.zeros(m)
    n=np.shape(S)[1]
    clusterCenter = np.mat(np.zeros((k,n)))
    for j in range(n):
        minJ=min(S[:,j])
        maxJ=max(S[:,j])
        rangeJ=float(maxJ-minJ)
        clusterCenter[:,j]=np.mat(minJ + rangeJ*np.random.rand(k,1))

    sampleTagChanged = True
    SSE = 0.0
    while sampleTagChanged:
        sampleTagChanged = False
        SSE = 0.0
        for i in range(m):
            minD = np.inf
            minIndex = -1
            for j in range(k):
                d=distMeas(clusterCenter[j,:],S[i,:])
                if d<minD:
                    minD=d
                    minIndex=j
            if sampleTag[i]!=minIndex:
                sampleTagChanged = True
            sampleTag[i] = minIndex
            SSE+=minD**2
        #print (SSE)
        for j in range(k):
            ClustI=S[np.nonzero(sampleTag[:]==j)[0]]
            clusterCenter[j,:]= np.mean(ClustI,axis=0)
            
    return clusterCenter,sampleTag,SSE    
def draw_pic(samples,clusterCenter,sampleTag):
    k=len(clusterCenter)
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    markers=['sg','py','ob','pk']
    for i in range(k):
        data_pos = samples[sampleTag== i]
        plt.plot(data_pos[:,0].tolist(),data_pos[:,1].tolist(),markers[i])
    plt.plot(clusterCenter[:,0].tolist(),clusterCenter[:,1].tolist(),"r*",markersize=20)
    plt.title('鸢尾花')
    plt.show()

def predict(datas,centroids,k):
    print  (type(centroids))
    return np.array([np.argmin(((np.tile(data,(k,1))-centroids)**2).sum(axis=1)) for data in datas])
 
def draw_range(data,centers,k):
    x_min,x_max=data[:,0].min()-1,data[:,0].max()+1
    y_min,y_max=data[:,1].min()-1,data[:,1].max()+1
    xx,yy = np.meshgrid(np.arange(x_min,x_max,0.02),np.arange(y_min,y_max,0.02))
    z = predict(np.c_[xx.ravel(),yy.ravel()],centers,k)
    z = z.reshape(xx.shape)
    cs = plt.contourf(xx,yy,z)

    
def main():
    k=4
    print ("----------ing-------------")

    choose = 0
    if choose == 0:
        iris_data = load_iris()
        data= iris_data.data[:,:2]
    else:
        data = np.genfromtxt("kmeans.txt",delimiter=" ")

    min_loss = 10000
    min_loss_clusterCenter = np.array([])
    min_loss_sampleTag = np.array([])

    for i in range(50):
        clusterCenter,sampleTag,sse = kMeans_way(data,k)
        loss = sse
        if loss <min_loss:
            min_loss =loss
            min_loss_clusterCenter = clusterCenter
            min_loss_sampleTag=sampleTag

    clusterCenter=np.array(min_loss_clusterCenter)
    sampleTag=min_loss_sampleTag
    draw_range(data,clusterCenter,k)
    draw_pic(data,clusterCenter,sampleTag)

main()

选择最合适的k值

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import load_iris
def distance(vex1,vex2):
    return np.sqrt(np.sum(np.power(vex1-vex2,2)))
def kMeans_way(S,k,distMeas=distance):
    m=np.shape(S)[0]
    sampleTag = np.zeros(m)
    n=np.shape(S)[1]
    clusterCenter = np.mat(np.zeros((k,n)))
    for j in range(n):
        minJ=min(S[:,j])
        maxJ=max(S[:,j])
        rangeJ=float(maxJ-minJ)
        clusterCenter[:,j]=np.mat(minJ + rangeJ*np.random.rand(k,1))
    sampleTagChanged = True
    SSE = 0.0
    while sampleTagChanged:
        sampleTagChanged = False
        SSE = 0.0
        for i in range(m):
            minD = np.inf
            minIndex = -1
            for j in range(k):
                d=distMeas(clusterCenter[j,:],S[i,:])
                if d<minD:
                    minD=d
                    minIndex=j
            if sampleTag[i]!=minIndex:
                sampleTagChanged = True
            sampleTag[i] = minIndex
            SSE+=minD**2
        #print (SSE)
        for j in range(k):
            ClustI=S[np.nonzero(sampleTag[:]==j)[0]]
            clusterCenter[j,:]= np.mean(ClustI,axis=0)
            
    return clusterCenter,sampleTag,SSE    
def draw_pic(samples,clusterCenter,sampleTag):
    k=len(clusterCenter)
    plt.rcParams['font.sans-serif']=['SimHei']
    plt.rcParams['axes.unicode_minus'] = False
    markers=['sg','py','ob','pk']
    for i in range(k):
        data_pos = samples[sampleTag== i]
        plt.plot(data_pos[:,0].tolist(),data_pos[:,1].tolist(),markers[i])
    plt.plot(clusterCenter[:,0].tolist(),clusterCenter[:,1].tolist(),"r*",markersize=20)
    plt.title('鸢尾花')
    plt.show()

def predict(datas,centroids,k):
    print  (type(centroids))
    return np.array([np.argmin(((np.tile(data,(k,1))-centroids)**2).sum(axis=1)) for data in datas])
 
def draw_range(data,centers,k):
    x_min,x_max=data[:,0].min()-1,data[:,0].max()+1
    y_min,y_max=data[:,1].min()-1,data[:,1].max()+1
    xx,yy = np.meshgrid(np.arange(x_min,x_max,0.02),np.arange(y_min,y_max,0.02))

    z = predict(np.c_[xx.ravel(),yy.ravel()],centers,k)
    print (xx.shape)
    print (z.shape)
    z = z.reshape(xx.shape)

    cs = plt.contourf(xx,yy,z)

    
def main():

    print ("----------ing-------------")
    data = np.genfromtxt("kmeans.txt", delimiter=" ")
    min_loss = 10000
    min_loss_clusterCenter = np.array([])
    min_loss_sampleTag = np.array([])
    loss_list =[]
    for k in range(2,10):
        clusterCenter,sampleTag,sse = kMeans_way(data,k)
        loss = sse
        if loss <min_loss:
            min_loss =loss
            min_loss_clusterCenter = clusterCenter
            min_loss_sampleTag=sampleTag
        loss_list.append(min_loss)

    print (loss_list)
    plt.plot(range(2,10),loss_list)
    plt.xlabel("x")
    plt.xlabel("loss")
    plt.show()

    """
    clusterCenter=np.array(min_loss_clusterCenter)
    sampleTag=min_loss_sampleTag
    draw_range(data,clusterCenter,k)
    draw_pic(data,clusterCenter,sampleTag)
    """

main()

说明：有明显拐点出，即为最佳值。（不一定都存在最佳点）

数据集（kmeans.txt）

1.658985 4.285136  
-3.453687 3.424321  
4.838138 -1.151539  
-5.379713 -3.362104  
0.972564 2.924086  
-3.567919 1.531611  
0.450614 -3.302219  
-3.487105 -1.724432  
2.668759 1.594842  
-3.156485 3.191137  
3.165506 -3.999838  
-2.786837 -3.099354  
4.208187 2.984927  
-2.123337 2.943366  
0.704199 -0.479481  
-0.392370 -3.963704  
2.831667 1.574018  
-0.790153 3.343144  
2.943496 -3.357075  
-3.195883 -2.283926  
2.336445 2.875106  
-1.786345 2.554248  
2.190101 -1.906020  
-3.403367 -2.778288  
1.778124 3.880832  
-1.688346 2.230267  
2.592976 -2.054368  
-4.007257 -3.207066  
2.257734 3.387564  
-2.679011 0.785119  
0.939512 -4.023563  
-3.674424 -2.261084  
2.046259 2.735279  
-3.189470 1.780269  
4.372646 -0.822248  
-2.579316 -3.497576  
1.889034 5.190400  
-0.798747 2.185588  
2.836520 -2.658556  
-3.837877 -3.253815  
2.096701 3.886007  
-2.709034 2.923887  
3.367037 -3.184789  
-2.121479 -4.232586  
2.329546 3.179764  
-3.284816 3.273099  
3.091414 -3.815232  
-3.762093 -2.432191  
3.542056 2.778832  
-1.736822 4.241041  
2.127073 -2.983680  
-4.323818 -3.938116  
3.792121 5.135768  
-4.786473 3.358547  
2.624081 -3.260715  
-4.009299 -2.978115  
2.493525 1.963710  
-2.513661 2.642162  
1.864375 -3.176309  
-3.171184 -3.572452  
2.894220 2.489128  
-2.562539 2.884438  
3.491078 -3.947487  
-2.565729 -2.012114  
3.332948 3.983102  
-1.616805 3.573188  
2.280615 -2.559444  
-2.651229 -3.103198  
2.321395 3.154987  
-1.685703 2.939697  
3.031012 -3.620252  
-4.599622 -2.185829  
4.196223 1.126677  
-2.133863 3.093686  
4.668892 -2.562705  
-2.793241 -2.149706  
2.884105 3.043438  
-2.967647 2.848696  
4.479332 -1.764772  
-4.905566 -2.911070