k-mean的代码怎么写

K-mean的代码实现可以参考以下步骤：
1. 随机选取k个点作为初始簇中心
2. 对于每个样本点，计算其与每个簇中心的距离，将其归入距离最近的簇
3. 对于每个簇，重新计算簇中心
4. 重复步骤2和3，直到簇中心不再发生变化或达到最大迭代次数

以下是Python的实现代码示例：

import numpy as np

def k_means(X, k, max_iterations):
    n_samples, n_features = X.shape

    # Initialize cluster centers
    centroids = X[np.random.choice(n_samples, k, replace=False)]

    # Iterate until convergence or max_iterations
    for i in range(max_iterations):
        # Assign points to nearest cluster
        clusters = [[] for _ in range(k)]
        for idx, sample in enumerate(X):
            distances = [np.linalg.norm(sample - centroid) for centroid in centroids]
            nearest_centroid_idx = np.argmin(distances)
            clusters[nearest_centroid_idx].append(idx)

        # Update centroid position
        new_centroids = np.zeros((k, n_features))
        for cluster_idx, cluster in enumerate(clusters):
            cluster_mean = np.mean(X[cluster], axis=0)
            new_centroids[cluster_idx] = cluster_mean

        # Check for convergence
        if np.allclose(centroids, new_centroids):
            break

        centroids = new_centroids

    return centroids, clusters