Pytorch 广播两个张量的乘积

Question

我想将两个张量相乘，这是我得到的：

A 形状的张量 (20, 96, 110)
B 形状的张量 (20, 16, 110)

第一个索引是批量大小。我想做的基本上是从 B - (20, 1, 110) 中获取每个张量，例如，我想乘以每个 A 张量 (20, n, 110)。所以产品将在最后：张量 AB 其形状为 (20, 96 * 16, 110).

所以我想通过 B 广播来乘以 A 中的每个张量。 PyTorch 中是否有方法可以做到这一点？

Answer 1

使用torch.einsum followed by torch.reshape：

AB = torch.einsum("ijk,ilk->ijlk", (A, B)).reshape(A.shape[0], -1, A.shape[2])

示例：

import numpy as np
import torch

# A of shape (2, 3, 2):
A = torch.from_numpy(np.array([[[1, 1], [2, 2], [3, 3]], 
                               [[4, 4], [5, 5], [6, 6]]]))
# B of shape (2, 2, 2):
B = torch.from_numpy(np.array([[[1, 1], [10, 10]], 
                               [[2, 2], [20, 20]]]))

# AB of shape (2, 3*2, 2):
AB = torch.einsum("ijk,ilk->ijlk", (A, B)).reshape(A.shape[0], -1, A.shape[2])
# tensor([[[ 1, 1], [ 10, 10], [  2,  2], [ 20,   20], [ 3,   3], [ 30,  30]],
#         [[ 8, 8], [ 80, 80], [ 10, 10], [ 100, 100], [ 12, 12], [ 120, 120]]])

Pytorch 广播两个张量的乘积

Pytorch broadcasting product of two tensors

python

matrix-multiplication

deep-learning

pytorch

tensor