不调用工具包，实现Kmeans聚类，改变不同的K值和距离度量方式，可视化聚类结果

以下是Python实现Kmeans聚类的代码，并可视化聚类结果：

import numpy as np
import matplotlib.pyplot as plt

class KMeans:
    def __init__(self, k=2, max_iter=100, distance_measure='euclidean'):
        self.k = k
        self.max_iter = max_iter
        self.distance_measure = distance_measure

    def fit(self, X):
        self.centroids = X[np.random.choice(X.shape[0], size=self.k, replace=False)]
        for i in range(self.max_iter):
            clusters = [[] for _ in range(self.k)]
            for x in X:
                distances = []
                for c in self.centroids:
                    if self.distance_measure == 'euclidean':
                        distances.append(np.linalg.norm(x - c))
                    elif self.distance_measure == 'manhattan':
                        distances.append(np.sum(np.abs(x - c)))
                cluster_index = np.argmin(distances)
                clusters[cluster_index].append(x)
            new_centroids = []
            for i in range(self.k):
                new_centroid = np.mean(clusters[i], axis=0)
                new_centroids.append(new_centroid)
            if np.allclose(self.centroids, new_centroids):
                break
            self.centroids = new_centroids

    def predict(self, X):
        y_pred = []
        for x in X:
            distances = []
            for c in self.centroids:
                if self.distance_measure == 'euclidean':
                    distances.append(np.linalg.norm(x - c))
                elif self.distance_measure == 'manhattan':
                    distances.append(np.sum(np.abs(x - c)))
            cluster_index = np.argmin(distances)
            y_pred.append(cluster_index)
        return np.array(y_pred)

# 生成数据
np.random.seed(0)
X = np.vstack((np.random.randn(100, 2) + np.array([2, 2]), np.random.randn(100, 2) + np.array([-2, -2]), np.random.randn(100, 2) + np.array([2, -2])))
y = np.hstack((np.zeros(100), np.ones(100), np.ones(100) * 2))

# 可视化原始数据
plt.scatter(X[:, 0], X[:, 1], c=y)
plt.title('Original Data')
plt.show()

# Kmeans聚类
kmeans = KMeans(k=3, max_iter=100, distance_measure='euclidean')
kmeans.fit(X)
y_pred = kmeans.predict(X)

# 可视化聚类结果
plt.scatter(X[:, 0], X[:, 1], c=y_pred)
plt.scatter(kmeans.centroids[:, 0], kmeans.centroids[:, 1], marker='x', s=200, linewidths=3, color='r')
plt.title('Kmeans Clustering Result')
plt.show()

上述代码中生成了一个包含三个簇的数据集，并且使用Kmeans聚类将其分为三个簇。可以通过改变K值和距离度量方式来尝试不同的聚类结果。