pytorch中expand的numpy等价物是什么？

wi3ka0sx 于 2022-12-13 发布在其他

关注(0)|答案(3)|浏览(165)

假设我有一个numpy数组x，形状为[1,5]，我想沿着0轴展开它，使得到的数组y的形状为[10，5]，并且对于每个i，y[i:i+1,:]等于x。
如果x是pytorchTensor，我可以简单地做

y = x.expand(10,-1)

但是numpy中没有expand，看起来像它的那些（expand_dims和repeat）看起来也不像它。
示例：

>>> import torch
>>> x = torch.randn(1,5)
>>> print(x)
tensor([[ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724]])
>>> print(x.expand(10,-1))
tensor([[ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724],
        [ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724]])

pytorch

来源：https://stackoverflow.com/questions/65234748/what-is-the-numpy-equivalent-of-expand-in-pytorch

3条答案

按热度按时间

mqkwyuun1#

你可以用np.broadcast_to来实现，但不能用负数：

>>> import numpy as np
>>> x = np.array([[ 1.3306,  0.0627,  0.5585, -1.3128, -1.4724]])
>>> print(np.broadcast_to(x,(10,5)))
[[ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]
 [ 1.3306  0.0627  0.5585 -1.3128 -1.4724]]

赞(0）回复(0）举报 2022-12-13

gorkyyrv2#

您可以使用np.tile，它会重复指定轴的元素，如下所示：

>>> x = np.range(5)
>>> x = np.expand_dims(x, 0)
>>> x.shape
(1, 5)
>>> y = np.tile(x, (10, 1))  # repeat axis=0 10 times and axis=1 1 time
>>> y.shape
(10, 5)

赞(0）回复(0）举报 2022-12-13

scyqe7ek3#

numpy拥有numpy.newaxis