KMeans算法

简介

KMeans算法是一种简单的分类算法，算法思想是先随机选N个点分别生成N个集合，然后把剩余点往集合归类。由于选择的质心并不是确定正确的，每次填充完数据需要更新质心，重新归类直到质心不再改变。。

python实现

import numpy as np
import random
import math
class XYZClusterer(object):
    def __init__(self, cluster_num:int):
        assert cluster_num > 0
        self.cluster_num = cluster_num
        self.X           = np.zeros(0)
        self.Y           = np.zeros(0)
        self.ndot        = 0
        self.un_ind      = np.zeros(0, dtype=np.int)        # unknown point index
        self.clas        = [[] for i in range(cluster_num)] # point index in cluster
        self.clas_ctx    = np.zeros(self.cluster_num)       # point center x
        self.clas_cty    = np.zeros(self.cluster_num)       # point center y

    def fit(self, X:np.ndarray, Y:np.ndarray):
        assert X.shape[0] == Y.shape[0] >= self.cluster_num
        self.X       = X
        self.