在本专栏前面几篇中曾记录了一下K-means的matlab代码，这次使用时发现并不好用，因此又整理了其他的K-means代码，实测可行。
matlab:
代码语言：javascript复制%% K-mens方法的matlab实现
%% 数据准备和初始化
clc
clear
x=[62,627;112,511;186,531;198,411;190,379;234,399;227,598;329,454;349,596;424,600;611,565;811,736;776,537;666,437;944,449;943,318;743,216;1076,252;899,178;995,91;1074,101;943,17;275,341];
z=zeros(2,2);
z1=zeros(2,2);
z=x(1:2,1:2);
%% 寻找聚类中心
while 1
count=zeros(2,1);
allsum=zeros(2,2);
for i=1:23 %对每一个样本i，计算到2个聚类中心的距离
temp1=sqrt((z(1,1)-x(i,1)).^2 (z(1,2)-x(i,2)).^2);
temp2=sqrt((z(2,1)-x(i,1)).^2 (z(2,2)-x(i,2)).^2);
if(temp1<temp2)
count(1)=count(1) 1;
allsum(1,1)=allsum(1,1) x(i,1);
allsum(1,2)=allsum(1,2) x(i,2);
else
count(2)=count(2) 1;
allsum(2,1)=allsum(2,1) x(i,1);
allsum(2,2)=allsum(2,2) x(i,2);
end
end
z1(1,1)=allsum(1,1)/count(1);
z1(1,2)=allsum(1,2)/count(1);
z1(2,1)=allsum(2,1)/count(2);
z1(2,2)=allsum(2,2)/count(2);
if(z==z1)
break;
else
z=z1;
end
end
%% 结果显示
disp(z1);%输出聚类终须
plot(x(:,1),x(:,2),'k*',...
'LineWidth',2,...
'MarkerSize',10,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.5,0.5,0.5])
hold on
plot(z1(:,1),z1(:,2),'ko',...
'LineWidth',2,...
'MarkerSize',10,...
'MarkerEdgeColor','k',...
'MarkerFaceColor',[0.5,0.5,0.5])
set(gca,'linewidth',2);
xlabel('x','fontsize',12);
ylabel('y','fontsize',12);代码效果：
不过这个只能实现2种聚类
python代码：
代码语言：javascript复制# -*- coding:utf-8 -*-
import numpy as np
from matplotlib import pyplot
class K_Means(object):
# k是分组数；tolerance‘中心点误差’；max_iter是迭代次数
def __init__(self, k=2, tolerance=0.0001, max_iter=300):
self.k_ = k
self.tolerance_ = tolerance
self.max_iter_ = max_iter
def fit(self, data):
self.centers_ = {}
for i in range(self.k_):
self.centers_[i] = data[i]
for i in range(self.max_iter_):
self.clf_ = {}
for i in range(self.k_):
self.clf_[i] = []
# print("质点:",self.centers_)
for feature in data:
# distances = [np.linalg.norm(feature-self.centers[center]) for center in self.centers]
distances = []
for center in self.centers_:
# 欧拉距离
# np.sqrt(np.sum((features-self.centers_[center])**2))
distances.append(np.linalg.norm(feature - self.centers_[center]))
classification = distances.index(min(distances))
self.clf_[classification].append(feature)
# print("分组情况:",self.clf_)
prev_centers = dict(self.centers_)
for c in self.clf_:
self.centers_[c] = np.average(self.clf_[c], axis=0)
# '中心点'是否在误差范围
optimized = True
for center in self.centers_:
org_centers = prev_centers[center]
cur_centers = self.centers_[center]
if np.sum((cur_centers - org_centers) / org_centers * 100.0) > self.tolerance_:
optimized = False
if optimized:
break
def predict(self, p_data):
distances = [np.linalg.norm(p_data - self.centers_[center]) for center in self.centers_]
index = distances.index(min(distances))
return index
if __name__ == '__main__':
x = np.array([[149, 663], [404, 707], [743, 754], [170, 511], [520, 490], [912, 500], [303, 287], [810, 246], [1011, 298], [653, 33]])
k_means = K_Means(k=2)
k_means.fit(x)
print(k_means.centers_)
for center in k_means.centers_:
pyplot.scatter(k_means.centers_[center][0], k_means.centers_[center][1], marker='*', s=150)
for cat in k_means.clf_:
for point in k_means.clf_[cat]:
pyplot.scatter(point[0], point[1], c=('r' if cat == 0 else 'b'))
predict = [[2, 1], [6, 9]]
for feature in predict:
cat = k_means.predict(predict)
pyplot.show()修改k值即可实现聚几类，不过只能实现1，2
更多类的聚类有待后续挖掘…
