240718 均值漂移聚类

结果

均值漂移：这些“模式”就对应于一群群局部最密集（local maxima） 分布的点。 均值
漂移算法的优点是它不需要事先确定集群的数量。

代码

# -*- coding: utf-8 -*-
import numpy as np
from sklearn.cluster import MeanShift, estimate_bandwidth

import utilities

# 加载数据
X = utilities.load_data('data_multivar.txt')

# 带宽估计
bandwidth = estimate_bandwidth(X, quantile=0.1, n_samples=len(X))

# 计算均值
meanshift_estimator = MeanShift(bandwidth=bandwidth, bin_seeding=True)
meanshift_estimator.fit(X)
labels = meanshift_estimator.labels_
centroids = meanshift_estimator.cluster_centers_
num_clusters = len(np.unique(labels))

print "Number of clusters in input data =", num_clusters

###########################################################
# 画出节点

import matplotlib.pyplot as plt
from itertools import cycle

plt.figure()

# 为不同的簇指定标记形状
markers = '.*xv'

for i, marker in zip(range(num_clusters), markers):
    # plot the points belong to the current cluster
    plt.scatter(X[labels==i, 0], X[labels==i, 1], marker=marker, color='k')

    # plot the centroid of the current cluster
    centroid = centroids[i]
    plt.plot(centroid[0], centroid[1], marker='o', markerfacecolor='k',
             markeredgecolor='k', markersize=15)

plt.title('Clusters and their centroids')
plt.show()