unbind¶

dragon.vm.torch.unbind(
  input,
  dim=0,
  copy=True
)[source]¶

Unpack input into chunks along the given dimension.

The number of outputs is equal to the size of dimension:

x = torch.tensor([[1, 2, 3], [4, 5, 6]])
# Shape: (2, 3) -> (3,), (3,)
print(torch.unbind(x, dim=0))

dim can be negative:

x = torch.tensor([[1, 2, 3], [4, 5, 6]])
# Shape: (2, 3) -> (2,), (2,), (2,)
print(torch.unbind(x, dim=-1))

Parameters:

• input (dragon.vm.torch.Tensor) – The input tensor.
• dim (int, optional, default=0) – The dimension to unpack.
• copy (bool, optional, default=True) – Copy or create the views of input.

Returns:

Sequence[dragon.vm.torch.Tensor] – The output tensors.