简单的kmeans算法解析

#-*- coding:utf-8 -*-
import numpy as np
def kmeas(x,k,maxIt):#将数据导入处理并找好随机k中心点
    numpoints,numDim=x.shape
    dataset=np.zeros((numpoints,numDim+1))
    dataset[:,:-1]=x#将x的都赋给dataset的除了最后一列
    centroids=dataset[np.random.randint(numpoints,size=k),:]#随机产生k个中心点
    centroids=dataset[0:2,:]#在0到2中随机产生中心点
    centroids[:,-1]=range(1,k+1)#初始化最后一行
    iterations=0
    oldcentroids=None
    while not shouldStop(oldcentroids,centroids,iterations,maxIt):
        print("iterations:\n",iterations)
        print("dataset:\n",dataset)
        print("centroids:",centroids)
        oldcentroids=np.copy(centroids)#copy就是产生相同的但是 下次变化一个互不影响
        iterations+=1
        updatellabels(dataset,centroids)
        centroids=getcentroids(dataset,k)
    return dataset
def shouldStop(oldcentroids,centroids,iterations,maxIt):#判断循环次数或者中心点有变化来停止
    if iterations>maxIt:
        return True
    return np.array_equal(oldcentroids,centroids)#equal自动把我们比较2个值是否一样一样返回true

def updatellabels(dataset,centroids):#划分每个点到中心点的距离以此分类为哪个类
    numpoints,nunDims=dataset.shape
    for i in range(0,numpoints):
        dataset[i,-1]=getlabelfromclosestcentroid(dataset[i,:-1],centroids)

def getlabelfromclosestcentroid(datasetrow,centroids):#比较每个点到中心点的距离
    label=centroids[0,-1]
    minDist=np.linalg.norm(datasetrow-centroids[0,:-1])#每个点到中心点的距离最小值
    for i in range(1,centroids.shape[0]):
        dist=np.linalg.norm(datasetrow-centroids[i,:-1])
        if dist<minDist:#如果小于最小距离将他赋予哪一类
            minDist=dist
            label=centroids[i,-1]#将改行点化为这一类
    print("minDist:",minDist)
    return label

def getcentroids(dataset,k):#更新中心点
    result=np.zeros((k,dataset.shape[1]))
    for i in range(1,k+1):
        onecluster=dataset[dataset[:,-1]==i,:-1]#取出一类中的点
        result[i-1,:-1]=np.mean(onecluster,axis=0)#按列求平均值
        result[i-1,-1]=i
    return result

x1=np.array([1,1])
x2=np.array([2,1])
x3=np.array([4,3])
x4=np.array([5,4])
textx=np.vstack((x1,x2,x3,x4))
result=kmeas(textx,2,10)
print("final result:",result)