Python实现K-means聚类算法并可视化生成动图步骤详解

获取m个n维的数据
随即选取k个点作为初始中心点
while keep_changing:
	for i in range(m):
		for j in range(k):
			计算每个点到center的距离
			判断离哪个点更近
		for center in range(k):
			更新类别中心点的坐标

import numpy as np
import matplotlib.pyplot as plt
import sklearn.datasets as datasets

def create_data():
    X,y = datasets.make_blobs(n_samples=1000,n_features=2,centers=[[1,0],[5,4],[2,3],[10,8],[7,4]])
    return X,y

def init_centers(data,k):
    m, n =data.shape
    # m 样本个数，n特征个数
    center_ids = np.random.choice(m,k)
    centers = data[center_ids]
    return centers

def cal_dist(ptA,ptB):
   return np.linalg.norm(ptA-ptB)

def kmeans_process(data,k):
    centers = init_centers(data, k)
    m, n = data.shape
    keep_changing = True
    pred_y = np.zeros((m,))

    while keep_changing:
        keep_changing = False
        # 计算剩余样本所属类别
        for i in range(m):
            min_distance = np.inf
            for center in range(k):
                distance = cal_dist(data[i,:],centers[center,:])
                if distance<min_distance: # 判断离哪个更近
                    min_distance = distance
                    idx = center # 类别换下
            if pred_y[i] != idx:   # 判断是否发生了改变
                keep_changing = True
            pred_y[i] = idx
        # 更新类别中心点坐标
        for center in range(k):
            cluster_data = data[pred_y==center]
            centers[center,:] = np.mean(cluster_data, axis=0) # 求相同类别数据点的质心点
        print(centers)
    return centers, pred_y

if __name__ == '__main__':
    X, y = create_data()
    centers , pred_y = kmeans_process(data=X, k=5)
    plt.scatter(X[:,0], X[:,1], s=3, c=pred_y)
    plt.scatter(centers[:,0], centers[:,1], s=10, c='k')
    plt.show()