K-means 聚类算法的 Python 实现示例

import numpy as np

class KMeans:
    def __init__(self, n_clusters=2, max_iter=300):
        self.n_clusters = n_clusters  # 聚类数目
        self.max_iter = max_iter  # 最大迭代次数

    def fit(self, X):
        # 初始化聚类中心
        self.centroids = X[np.random.choice(X.shape[0], self.n_clusters, replace=False)]
        
        for _ in range(self.max_iter):
            # 计算每个样本到聚类中心的距离
            distances = np.sqrt(((X - self.centroids[:, np.newaxis])**2).sum(axis=2))
            
            # 将样本分配到最近的聚类中心
            labels = np.argmin(distances, axis=0)
            
            # 更新聚类中心
            new_centroids = np.array([X[labels == i].mean(axis=0) for i in range(self.n_clusters)])
            
            # 判断聚类中心是否变化小于阈值
            if np.allclose(self.centroids, new_centroids):
                break
            else:
                self.centroids = new_centroids

        return labels

# 示例数据
X = np.array([[1, 2], [5, 8], [1.5, 1.8], [8, 8], [1, 0.6], [9, 11]])

# 聚类
kmeans = KMeans(n_clusters=2)
labels = kmeans.fit(X)

print("聚类结果：", labels)