3.K均值算法

from sklearn.datasets import load_iris
import numpy as np

iris = load_iris()
data = iris[‘data‘]
data.shape
n = len(data)   # 样本个数
m = data.shape[1]  # 样本属性个数
k = 3   # 类中心个数
dist = np.zeros([n, k+1])   # 初始化距离矩阵，最后一列存放每个样本的类别（归属的类）
# 选中心
center = data[:k, :]  # 选择前k个样本为初始类中心
center_new = np.zeros([k, m])  # 新的类中心
while True:
    for i in range(n):
        for j in range(k):
            dist[i, j] = np.sqrt(sum((data[i, :] - center[j, :])**2))  # 2、求距离
        dist[i, k] = np.argmin(dist[i, :k])  # 3、归类
    # 求新类中心
    for i in range(k):
        index = dist[:, k] == i  # 求类别为k的样本索引, 判断距离矩阵中最后一列归属为哪一类
        center_new[i,:] = data[index, :].mean(axis=0)  # 取得k类样本的属性值得均值，作为新的类中心
    # 判定结束
    if np.all((center == center_new)):
        break
    else:
        center = center_new
print(‘150个样本的归类：‘, dist[:, k])

import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.datasets import load_iris
iris=load_iris()
data=iris.data[:,1]
x=data.reshape(-1,1)
y=KMeans(n_clusters=3)
y.fit(x)
y_pre=y.predict(x)
plt.scatter(x[:,0],x[:,0],c=y_pre,s=50,cmap=‘rainbow‘)
plt.show()

import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.datasets import load_iris
data=load_iris()
x=data.data
y=KMeans(n_clusters=3)
y.fit(x)
y_pre=y.predict(x)
plt.scatter(x[:,2],x[:,3],c=y_pre,s=100,cmap=‘rainbow‘,alpha=1)
plt.show()