How to select index over two dimension in PyTorch?

user_1831512 0
Viewed 2915

Given a = torch.randn(3, 2, 4, 5), how I can select sub tensor like (2, :, 0, :), (1, :, 1, :), (2, :, 2, :), (0, :, 3, :) (a resulting tensor of size (2, 4, 5) or (4, 2, 5)?

While a[2, :, 0, :] gives 

 0.5580 -0.0337  1.0048 -0.5044  0.6784
-1.6117  1.0084  1.1886  0.1278  0.3739
[torch.FloatTensor of size 2x5]

however, a[[2, 1, 2, 0], :, [0, 1, 2, 3], :] gives

TypeError: Performing basic indexing on a tensor and encountered an error indexing dim 0 with an object of type list. The only supported types are integers, slices, numpy scalars, or if indexing with a torch.LongTensor or torch.ByteTensor only a single Tensor may be passed.

though numpy returns a (4, 2, 5) tensor successfully.

pytorch numpy
1 Answers

Does it work for you?

import torch

a = torch.randn(3, 2, 4, 5)
print(a.size())

b = [a[2, :, 0, :], a[1, :, 1, :], a[2, :, 2, :], a[0, :, 3, :]]
b = torch.stack(b, 0)

print(b.size()) # torch.Size([4, 2, 5])
user_53523990
Related