深度解读Python如何实现dbscan算法

from sklearn.cluster import DBSCAN
import numpy as np

# 创建样本数据
X = np.array([[1, 2], [2, 2], [2, 3], [8, 7], [8, 8], [25, 80]])

# 创建并训练模型
db = DBSCAN(eps=3, min_samples=2).fit(X)

# 获取聚类标签
labels = db.labels_

# 打印聚类结果
print("Labels:", labels)

import numpy as np

def euclidean_distance(x, y):
    return np.sqrt(np.sum((x - y)**2))

def dbscan(X, eps, min_samples):
    m = X.shape[0]
    labels = [0] * m
    C = 0
    for i in range(m):
        if labels[i] != 0:
            continue
        neighbors = []
        for j in range(m):
            if euclidean_distance(X[i], X[j]) < eps:
                neighbors.append(j)
        if len(neighbors) < min_samples:
            labels[i] = -1
        else:
            C += 1
            labels[i] = C
            for j in neighbors:
                labels[j] = C
    return labels

X = np.array([[1,2],[2,2],[2,3],[8,7],[8,8],[25,80]])
labels = dbscan(X, 3, 2)
print(labels)

import numpy as np
import matplotlib.pyplot as plt
from sklearn.datasets import make_moons
from sklearn.cluster import DBSCAN

# 创建数据集
X, y = make_moons(n_samples=200, noise=0.05, random_state=0)

# 初始化 DBScan 模型
dbscan = DBSCAN(eps=0.3, min_samples=5)

# 训练模型
y_pred = dbscan.fit_predict(X)

# 可视化结果
plt.scatter(X[:, 0], X[:, 1], c=y_pred)
plt.show()