numpy：选择多个算术序列元素以形成多个数组

Suppose we get an array using following code, what is the efficient way to get array[0], array[N], ..., array[MN-N] to form a new array, and get array[1], array[N+1],..., array[MN-N+1] to form another array, and ..., etc.

假设我们使用以下代码得到一个数组,有什么方法可以得到数组[0],数组[N],...,数组[MN-N]形成一个新数组,得到数组[1],数组[N + 1],...,数组[MN-N + 1]形成另一个数组,......等

array = []
for i in range(M):
    for j in range(N):
        array.append(something)

1 个解决方案

#1

With numpy, you can use reshape:

有了numpy,你可以使用reshape:

np.reshape(array, (M, N))

Then the columns are the arrays you are looking for.

然后列是您要查找的数组。

M = 3
N = 2

array = []
for i in range(M):
    for j in range(N):
        array.append(i+j)

array
# [0, 1, 1, 2, 2, 3]

np.reshape(array, (M, N))
# array([[0, 1],
#        [1, 2],
#        [2, 3]])

#1