如何修复标量变量的无效索引？答案

【问题标题】：How to fix Invalid index to scalar variable?如何修复标量变量的无效索引？
【发布时间】：2020-09-13 00:53:36
【问题描述】：

我尝试编写 K-means 算法并在行中遇到此错误：points = np.array ([x[j] for j in range (len (x)) if cluster [j] == i]) 有人可以帮忙吗？

from copy import deepcopy

def euclidean_distance (a, b, ax = 1):
    return np.linalg.norm (a - b , axis = ax)

c_prev = np.zeros (c.shape)
clusters = np.zeros (len(x))

distance_differences = euclidean_distance (c, c_prev)

while distance_differences.any () != 0:
    for i in range (len(x)):
        distances = euclidean_distance (x[i], c)
        cluster = np.argmin (distances)
        clusters [i] = cluster
    c_prev = deepcopy (c)
    for i in range (k):
        points = [x[j] for j in range (len(x)) if clusters [j] == i]
        if len(points) != 0:
            c[i] = np.mean (points, axis = 0)

    distance_differences = euclidean_distance (c, c_prev)

colors = ['b', 'r', 'y', 'g', 'c', 'm']
for i in range (k):
    points = np.array ([x[j] for j in range (len (x)) if cluster [j] == i])
    if len(points) > 0:
        plt.scatter (points [:, 0], points [:, 1], s= 10, c = colors [i])
    else:
        print ('Please regenerate your centeroids again')

plt.scatter (points [:, 0], points [:, 1], s= 10, c = colors [i])
plt.scatter(c[:, 0], c[:, 1], marker = '*', s =100, c ='k')
plt.show();

【问题讨论】：

请详细说明错误并可能解释您要达到的目标，而不是期望读者事先知道您的意思！
还有c、x和k是什么？
抱歉简短的解释。我想编写 k-means 聚类算法并对其进行测试。 c 是随机质心，x 是包含所有示例点的矩阵，k 是簇的编号。

标签： python-3.x numpy jupyter-notebook k-means

【解决方案1】：

是的，你的行中有错字，

points = np.array ([x[j] for j in range (len (x)) if cluster [j] == i])

应该是这样，

points = np.array ([x[j] for j in range (len (x)) if clusters[j] == i])

因为cluster 只是一个标量（即只有一个值）&clusters 是数组

【讨论】：

是的，这是我的错误，现在可以使用了！谢谢！