I have:
print(f"inp: {inp.size()}\tleft_end_indices: {left_end_indices.size()}\tself.side_length: {self.side_length}")
left_remain_input = inp[:,left_end_indices + 1:left_end_indices + 1 + self.side_length]
which has the sizes:
inp: torch.Size([16, 1544, 161]) left_end_indices: torch.Size([16]) self.side_length: 48
But the error I get:
TypeError: only integer tensors of a single element can be converted to an index
Basically, I want to set left_remain_input
to be portions of inp
based on where the left_end_indices
are in the second dimension. How can I accomplish this with PyTorch?