weighted k-means代码

下面是使用Python实现的weighted k-means算法的示例代码：

import numpy as np
from scipy.spatial.distance import cdist

def weighted_kmeans(X, k, weights, max_iterations=100):
    # Initialize centroids randomly
    centroids = X[np.random.choice(X.shape[0], k, replace=False)]
    
    for i in range(max_iterations):
        # Calculate distances between data points and centroids
        distances = cdist(X, centroids, 'euclidean')
        
        # Assign data points to nearest centroid based on weighted distances
        cluster_ids = np.argmin(distances * weights[:, None], axis=1)
        
        # Update centroids based on weighted mean of assigned data points
        for j in range(k):
            centroid = np.mean(X[cluster_ids == j] * weights[cluster_ids == j][:, None], axis=0)
            centroids[j] = centroid
        
    return centroids, cluster_ids

这个函数接收三个参数：

- X: 数据矩阵，每一行代表一个数据点。
- k: 聚类的数量。
- weights: 每个数据点的权重，应该是一个长度为X.shape[0]的向量。

函数返回两个值：

- centroids: 聚类中心的坐标矩阵，每一行代表一个中心。
- cluster_ids: 每个数据点所属的聚类的ID，应该是一个长度为X.shape[0]的向量。