我想在选定的轴上用0填充numpy张量。
例如,我有张量r
形状(4,3,2)
,但我只对填充最后两个轴(即仅填充矩阵)感兴趣。是否可以使用单行python代码?
答案 0 :(得分:56)
您可以使用np.pad()
:
a = np.ones((4, 3, 2))
# npad is a tuple of (n_before, n_after) for each dimension
npad = ((0, 0), (1, 2), (2, 1))
b = np.pad(a, pad_width=npad, mode='constant', constant_values=0)
print(b.shape)
# (4, 6, 5)
print(b)
# [[[ 0. 0. 0. 0. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0.]]
# [[ 0. 0. 0. 0. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0.]]
# [[ 0. 0. 0. 0. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0.]]
# [[ 0. 0. 0. 0. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 1. 1. 0.]
# [ 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0.]]]
答案 1 :(得分:2)
此功能将在某个轴的末尾填充 如果你想填两面,只需修改它。
def pad_along_axis(array: np.ndarray, target_length, axis=0):
pad_size = target_length - array.shape[axis]
axis_nb = len(array.shape)
if pad_size < 0:
return a
npad = [(0, 0) for x in range(axis_nb)]
npad[axis] = (0, pad_size)
b = np.pad(array, pad_width=npad, mode='constant', constant_values=0)
return b
示例:
>>> a = np.identity(5)
>>> b = pad_along_axis(a, 7, axis=1)
>>> print(a,a.shape)
[[1. 0. 0. 0. 0.]
[0. 1. 0. 0. 0.]
[0. 0. 1. 0. 0.]
[0. 0. 0. 1. 0.]
[0. 0. 0. 0. 1.]] (5, 5)
>>> print(b,b.shape)
[[1. 0. 0. 0. 0. 0. 0.]
[0. 1. 0. 0. 0. 0. 0.]
[0. 0. 1. 0. 0. 0. 0.]
[0. 0. 0. 1. 0. 0. 0.]
[0. 0. 0. 0. 1. 0. 0.]] (5, 7)