def test_binary_k_means():
k = 3
data_mat = mat(kMeans.load_data_set('testSet2.txt'))
center, cluster = kMeans.binary_k_means(data_mat, k)
kMeans.plot_k_means(data_mat, cluster, center, k)
def test_load():
data_mat = mat(kMeans.load_data_set('testSet.txt'))
print min(data_mat[:, 0])
print max(data_mat[:, 1])
print kMeans.rand_center(data_mat, 2)
def test_k_means():
k = 4
data_mat = mat(kMeans.load_data_set('testSet.txt'))
center, cluster = kMeans.k_means(data_mat, k)
kMeans.plot_k_means(data_mat, cluster, center, k)
"""
轮廓系数相关代码
"""
import numpy
from sklearn.cluster import KMeans
from sklearn.metrics import silhouette_score
import matplotlib.pyplot as plt
from kMeans import load_data_set
if __name__ == '__main__':
data_mat = numpy.mat(load_data_set('testSet2.txt'))
# 分辨率1280 1024
plt.figure(figsize=(12.8, 10.24))
# 分割出3*2=6个子图，并在1号作图
plt.subplot(3, 2, 1)
x1_min = min(data_mat[:, 0]) - 1
x1_max = max(data_mat[:, 0]) + 1
x2_min = min(data_mat[:, 0]) - 1
x2_max = max(data_mat[:, 1]) + 1
plt.xlim([x1_min, x1_max])
plt.ylim([x2_min, x2_max])
plt.title('Instance')
plt.scatter(data_mat[:, 0].tolist(), data_mat[:, 1].tolist())
colors = 'bgrcmykb'
markers = 'osDv^p*+'
clusters = [2, 3, 4, 5, 8]