如何使用每个集群独特的颜色可视化输出集群

Question

我只是python中的新手

我在互联网上搜索执行 K-means 使用 scikit 的代码，我尝试修改代码以可视化 plot 3d 并为每个集群（3 个集群）着色，但结果是所有集群具有相同颜色，代码和如下图：

import numpy as np
import matplotlib.pyplot as plt
from matplotlib import style
style.use("ggplot")
from sklearn.cluster import KMeans
from collections import Counter
from mpl_toolkits.mplot3d import Axes3D
from pylab import *

X = np.array([[1, 2, 5],
              [5, 8, 2],
              [1.5, 1.8, 6],
              [8, 8, 9],
              [1, 0.6, 10],
              [2.5, 3.8, 6],
              [2.5, 5.8, 9],
              [5, 8, 3],
              [4, 0.6, 7],
              [2.5, 1.8, 4.6],
              [6.5, 1.8, 12],
              [7, 8, 9],
              [2, 0.6, 7],
              [5.5, 1.8, 4],
              [4.8, 6.9, 6],
              [4.9, 9.8, 2],
              [9, 11, 12]])


cluster_num = 3

kmeans = KMeans(n_clusters=cluster_num)
kmeans.fit(X)

centroids = kmeans.cluster_centers_
labels = kmeans.labels_

print "centroids : "
print centroids
print "labels : "
print labels

colors = ["g.","r.","c.","y."]

color = np.random.rand(cluster_num)

c = Counter(labels)


fig = figure()
ax = fig.gca(projection='3d')


for i in range(len(X)):
    print("coordinate:",X[i], "label:", labels[i])
    print "i : ",i
    print "color[labels[i]] : ",color[labels[i]]
    ax.scatter(X[i][0], X[i][1], X[i][2], c=color[labels[i]])


for cluster_number in range(cluster_num):
  print("Cluster {} contains {} samples".format(cluster_number, c[cluster_number]))

ax.scatter(centroids[:, 0],centroids[:, 1], centroids[:, 2], marker = "x", s=150, linewidths = 5, zorder = 100)

plt.show()

我怎样才能让每个集群都有自己的颜色？谢谢

Answer 1

现在color = np.random.rand(cluster_num) 正在生成三个随机数，而在ax.scatter(X[i][0], X[i][1], X[i][2], c=color[labels[i]]) 中，您正试图将这些随机数分配为颜色。

相反，您可以更改color = ["g", "r", "b"]，使第一个集群为绿色，第二个为红色，第三个为蓝色。

对于聚类中心，传递相同的参数：

ax.scatter(centroids[:, 0],centroids[:, 1], centroids[:, 2], marker = "x", s=150, linewidths = 5, zorder = 100, c=color)