python - 如何获得 numpy 数组中非零值的位置？

Question

我正在使用 np.nonzero() 但我不明白返回值

我试试

for groupPosition in np.nonzero(groupMatrix):
    print groupPosition

并返回 [0 0 1 2 3 3 3]

for groupPosition in zip(np.nonzero(groupMatrix)):
    print groupPosition

并返回 (array([0, 1, 0, 3, 0, 1, 3]),)

组矩阵:

[[ 1.  1.  0.  0.]
[ 1.  0.  0.  0.]
[ 0.  0.  0.  2.]
[ 3.  3.  0.  2.]]

但不要像 (0, 0) 那样返回位置

Answer 1

>>> import numpy as np
>>>
>>> groupMatrix = np.array([
...     [1, 1, 0, 0],
...     [1, 0, 0, 0],
...     [0, 0, 0, 2],
...     [3, 3, 0, 2]
... ])
>>> np.nonzero(groupMatrix)
(array([0, 0, 1, 2, 3, 3, 3], dtype=int64), array([0, 1, 0, 3, 0, 1, 3], dtype=int64))
>>> zip(np.nonzero(groupMatrix))
[(array([0, 0, 1, 2, 3, 3, 3], dtype=int64),), (array([0, 1, 0, 3, 0, 1, 3], dtype=int64),)]

使用zip(*...):

>>> zip(*np.nonzero(groupMatrix))
[(0, 0), (0, 1), (1, 0), (2, 3), (3, 0), (3, 1), (3, 3)]

---

zip(*a) 类似于 zip(a[0], a[1], ...)

>>> a = [(0, 1, 2), (3, 4, 5)]
>>> zip(a)
[((0, 1, 2),), ((3, 4, 5),)]
>>> zip(a[0], a[1])
[(0, 3), (1, 4), (2, 5)]
>>> zip(*a)
[(0, 3), (1, 4), (2, 5)]

参见 Unpacking Argument Lists .