Map each element of torch.Tensor with it's value in the dict

Question

Suppose i have a tensor t consisting only zeros and ones:

t = torch.Tensor([1, 0, 0, 1])

And a dict with the weights:

weights = {0: 0.1, 1: 0.9}

I want to form a new tensor new_t, such that every element in tensor t is mapped to the corresponding value in the dict weights:

new_t = torch.Tensor([0.9, 0.1, 0.1, 0.9])

Is there an elegant way to do this without iterating over tensor t? I've heard about torch.apply, but it only works if tensor t is on the CPU, is there any other options?

Karl · Accepted Answer · 2025-01-30 18:43:13Z

1

If you convert your weights dict into a tensor, you can index directly

t = torch.tensor([1, 0, 0, 1])
weights = torch.tensor([0.1, 0.9])

new_t = weights[t]
new_t
>tensor([0.9000, 0.1000, 0.1000, 0.9000])

answered Jan 30 at 18:43

Karl

5,9661 gold badge11 silver badges19 bronze badges

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

1 Comment

onthebox Jan 30 at 21:25

Thanks! I didn't know you can index a tensor this way

Collectives™ on Stack Overflow

Map each element of torch.Tensor with it's value in the dict

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related