PyTorch - Tensors multiplication along new dimension

Question

Sorry if already asked, but I can't find the words to look for on Google.

Let's say I have a tensor t1 of size [a,b] and a tensor t2 of size [c].

How can I output a tensor t3 of size [a,b,c], so that:

t3[0, :, 0] = t1[0, :] * t2[0] t3[0, :, 1] = t1[0, :] * t2[1] t3[0, :, 2] = t1[0, :] * t2[2] ... t3[1, :, 0] = t1[1, :] * t2[0] t3[1, :, 1] = t1[1, :] * t2[1] t3[1, :, 2] = t1[1, :] * t2[2]

and so on, without a for loop ?

Thanks

mpromonet · Accepted Answer · 2021-06-13 17:45:32Z

2

Using torch.kron will give a tensor a x b*c, then using torch.Tensor.reshape could map to a tensor a x b x c :

t3=torch.kron(t1,t2).reshape(a,b,c)

edited Jun 13, 2021 at 17:45

answered Jun 11, 2021 at 19:34

mpromonet

12.1k45 gold badges68 silver badges99 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

PyTorch - Tensors multiplication along new dimension

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related