python - 为numpy数组切片的奇数和偶数元素分配不同的值

Question

是否可以为 numpy 数组切片的偶数和奇数元素分配不同的值？

将起作用的代码示例：

a=np.zeros(100)
b=np.arange(10,50,10)

print(a)
[0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
 0. 0. 0. 0.]

a[b]=10

print(a)

[ 0.  0.  0.  0.  0.  0.  0.  0.  0.  0. 10.  0.  0.  0.  0.  0.  0.  0.
  0.  0. 10.  0.  0.  0.  0.  0.  0.  0.  0.  0. 10.  0.  0.  0.  0.  0.
  0.  0.  0.  0. 10.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.
  0.  0.  0.  0.  0.  0.  0.  0.  0.  0.]

现在，我正在尝试为切片 a[b] 的偶数和奇数元素分配不同的数字

a[b][::2] = 10

a[b][1::2] = -10

但这不起作用，初始的零数组不会修改。

简短的通知-实际上，b和a是动态生成的，因此我通常不知道它们的形状（a和b）以及哪些索引将在（b）内

标签： pythonnumpy

Answer 1

您应该直接索引b：

import numpy as np

a = np.zeros(100)
b = np.arange(10,50,10)

a[b[::2]] = 10
a[b[1::2]] = -10

print(a[b])
# [ 10. -10.  10. -10.]