第八次做業：聚類--K均值算法：自主實現與sklearn.cluster.KMeans調用

import numpy as np
x = np.random.randint(1,100,[20,1])
y = np.zeros(20)
k = 3
def initcenter(x,k):
    return x[:k]
 
kc = initcenter(x,k)
kc
def nearest(kc,i):
    d=(abs(kc-i))
    w=np.where(d==np.min(d))
    return w[0][0]
 
kc = initcenter(x,k)
nearest(kc,56)
def xclassify(x,y,kc):
    for i in range(x.shape[0]):
        y[i] = nearest(kc,x[i])
    return y
 
kc = initcenter(x,k)
y = xclassify(x,y,kc)
print(kc,y)
def kcmean(x,y,kc,k):
    l = list(kc)
    flag = False
    for c in range(k):
        m = np.where(y == c)
        n = np.mean(x[m])
        if l[c] != n:
            l[c] = n
            flag = True
            print(l,flag)
    return (np.array(l),flag)
 
kc = initcenter(x,k)
flag = True
k = 3
 
 
while flag:
    y = xclassify(x,y,kc)
    kc,flag = kcmean(x,y,kc,k)

運行結果

 

二.鳶尾花花瓣長度數據作聚類並用散點圖顯示

import numpy as np
from sklearn.datasets import load_iris    
iris = load_iris()
x = iris.data[:,1]
y = np.zeros(150)

def initcenter(x,k):    #初始聚類中心數組
    return x[0:k].reshape(k)

def nearest(kc,i):       #數組中的值，與聚類中心最小距離所在類別的索引號
    d = (abs(kc-i))
    w = np.where(d == np.min(d))
    return w[0][0]

def xclassify(x,y,kc):
    for i in range(x.shape[0]):       #對數組的每一個值進行分類，shape[0]讀取矩陣第一維度的長度
        y[i] = nearest(kc,x[i])
    return y

def kcmean(x,y,kc,k):     #計算各聚類新均值
    l = list(kc)
    flag = False
    for c in range(k):
        print(c)
        m = np.where(y == c)
        n=np.mean(x[m])
        if l[c] != n:
            l[c] = n
            flag = True     #聚類中心發生變化
            print(l,flag)
    return (np.array(l),flag)


k = 3
kc = initcenter(x,k)

flag = True
print(x,y,kc,flag)

#判斷聚類中心和目標函數的值是否發生改變，若不變，則輸出結果，若改變，則返回2
while flag:
    y = xclassify(x,y,kc)
    kc, flag = kcmean(x,y,kc,k)
    print(y,kc,type(kc))
    
print(x,y)
import matplotlib.pyplot as plt
plt.scatter(x,x,c=y,s=50,cmap="rainbow");
plt.show()

運行結果