python sklearn 聚类 DBSCAN算法

文件下载地址：
链接: https://pan.baidu.com/s/1JB-h5dAGdh5bzfckaG7k5w 提取码: x62m

"""
    聚类 DBSCAN算法
    做圆法计算 外周样本 孤立样本 和核心样本
"""
import numpy as np
import sklearn.cluster as sc
import matplotlib.pyplot as mp
import sklearn.metrics as sm

x = np.loadtxt("./perf.txt", delimiter=",")
print(x.shape)

# 循环遍历 优选半径参数
eps, scores, models = np.linspace(0.3, 1.2, 10), [], []

for r in eps:
    # DBSCAN算法 eps 半径 min_samples 最小样本数
    model = sc.DBSCAN(eps=r, min_samples=5)
    model.fit(x)
    labels = model.labels_
    # 计算轮廓系数
    score = sm.silhouette_score(x, labels, sample_size=len(x), metric="euclidean")
    scores.append(score)
    models.append(model)
scores = np.array(scores)
models = np.array(models)

# 获取最优得分 最优模型以及分数
best_index = np.argmax(scores)

best_model = models[best_index]
best_score = scores[best_index]
best_r = eps[best_index]
print("最优分数:", best_score, "最优半径:", best_r)
print("最佳模型:\n", best_model)

# 绘制图像
mp.figure('DBSCAN Cluster', facecolor='lightgray')
mp.title('DBSCAN Cluster', fontsize=20)
mp.xlabel('x', fontsize=14)
mp.ylabel('y', fontsize=14)
mp.tick_params(labelsize=10)
labels = best_model.labels_

# 核心样本
core_mask = np.zeros(len(x), dtype="bool")
core_mask[best_model.core_sample_indices_] = True
mp.scatter(x[core_mask][:, 0], x[core_mask][:, 1], c=labels[core_mask], cmap="jet", s=70)

# 孤立样本
offset_mask = labels == -1
mp.scatter(x[offset_mask][:, 0], x[offset_mask][:, 1],
           marker="D", color="black", s=70, alpha=0.4)

# 外周样本
p_mask = ~(core_mask | offset_mask)
mp.scatter(x[p_mask][:, 0], x[p_mask][:, 1],
           marker="s", c=labels[p_mask], cmap="jet", s=70, alpha=0.4)

mp.show()