K 表示聚类:更新每个聚类的质心并选择颜色的函数

Question

这是我正在经历的K均值聚类示例的摘录。有人可以帮助我了解最后两行发生了什么吗？

具体说来：

1. class_of_points = compare_to_first_center > compare_to_second_center在做什么？它只是返回一个布尔值吗？
2. 同样在下一行中，colors_map[class_of_points + 1 - 1]在做什么？

提前谢谢，伙计们。

import random # library for random number generation
import numpy as np # library for vectorized computation
import pandas as pd # library to process data as dataframes
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans 
from sklearn.datasets.samples_generator import make_blobs
# data
x1 = [-4.9, -3.5, 0, -4.5, -3, -1, -1.2, -4.5, -1.5, -4.5, -1, -2, -2.5, -2, -1.5, 4, 1.8, 2, 2.5, 3, 4, 2.25, 1, 0, 1, 2.5, 5, 2.8, 2, 2]
x2 = [-3.5, -4, -3.5, -3, -2.9, -3, -2.6, -2.1, 0, -0.5, -0.8, -0.8, -1.5, -1.75, -1.75, 0, 0.8, 0.9, 1, 1, 1, 1.75, 2, 2.5, 2.5, 2.5, 2.5, 3, 6, 6.5]
#Define a function that updates the centroid of each cluster
colors_map = np.array(['b', 'r'])
def assign_members(x1, x2, centers):
compare_to_first_center = np.sqrt(np.square(np.array(x1) - centers[0][0]) + np.square(np.array(x2) - centers[0][1]))
compare_to_second_center = np.sqrt(np.square(np.array(x1) - centers[1][0]) + np.square(np.array(x2) - centers[1][1]))
class_of_points = compare_to_first_center > compare_to_second_center
colors = colors_map[class_of_points + 1 - 1]
return colors, class_of_points

Answer 1

>compare_to_first_center是所有点到centers[0]的距离，同样，compare_to_second_center是所有点到centers[1]的距离。现在，class_of_points是一个与您的点大小相同的布尔数组，说明每个点是否更接近center[0]或centers[1]。如果class_of_points[i]True，则数据中的point[i]更接近centers[0]。

colors = colors_map[class_of_points + 1 - 1]为点分配颜色b或r，b它们是否更接近centers[1]，并为centers[0]分配r。请注意，为了将布尔掩码class_of_points转换为索引数组，它们加 1 并减去 1，以便输出将False转换为0和True转换为 1，这使它们成为索引。一个例子是：

np.array([True, False, True])+1-1 

与

[1, 0, 1]

或者，您可以简单地将其替换为：

colors = colors_map[class_of_points + 0]

Answer 2

似乎给出了两个中心列表。此代码将计算每个点到每个中心点的欧氏距离，并将蓝色分配给centers[0][:]中靠近中心点的点，将红色分配给centers[1][:]中靠近中心点的点。

def assign_members(x1, x2, centers):
# In the following two lines, the eucledean distances are being calculated
compare_to_first_center = np.sqrt(np.square(np.array(x1) - centers[0][0]) + np.square(np.array(x2) - centers[0][1]))
compare_to_second_center = np.sqrt(np.square(np.array(x1) - centers[1][0]) + np.square(np.array(x2) - centers[1][1]))
# Check which center is closer to each point
# So class_of_points is a binary arary specifies the class number
class_of_points = compare_to_first_center > compare_to_second_center
# Depending on the class number (i.e. 0 or 1) it chooses the colour (blue or red)
colors = colors_map[class_of_points + 1 - 1]
return colors, class_of_points