Python张量矩阵乘法 [英] Python tensor matrix multiply

问题描述

我有张量

A = 
[[[a,b],
  [c,d]],
 [[e,f],
  [g,h]]]

和矩阵

B = 
[[1,2],
 [3,4]]

我需要获得

C = 
[[a*1+e*2,b*1+f*2],
 [c*3+g*4,d*3+h*4]]

如何使用矩阵形式的numpy执行此操作?我已经调查过np.tensordot()，但在这种情况下似乎无济于事.

How can I do this using numpy in matrix form? I've looked into np.tensordot() but it doesn't seem to help in this case.

推荐答案

您可以尝试以下操作:

>>> import numpy as np
>>> a = np.arange(1,9).reshape(2,2,2)
>>> a
array([[[1, 2],
        [3, 4]],

       [[5, 6],
        [7, 8]]])
>>> b = np.arange(1,5).reshape(2,2)
>>> b
array([[1, 2],
       [3, 4]])
>>> (a * b[None,:,:].T).sum(axis = 0)
array([[11, 14],
       [37, 44]])

中间步骤如下:

>>> b[None,:,:]
array([[[1, 2],
        [3, 4]]])
>>> b[None,:,:].T
array([[[1],
        [3]],

       [[2],
        [4]]])

这篇关于Python张量矩阵乘法的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！