TensorDictBase.unbind#
- TensorDictBase.unbind(dim: int) tuple[TensorDictBase, ...]#
Returns a tuple of the tensordicts along a batch dimension, as
torch.Tensor.unbind()does.The tensordicts of the result hold views of the original tensors. They are locked if the tensordict is shared or memory-mapped.
- Parameters:
dim (int) – the batch dimension to remove. The result has
batch_size[dim]tensordicts.
Examples
>>> import torch >>> from tensordict import TensorDict >>> td = TensorDict({ ... 'x': torch.arange(12).reshape(3, 4), ... }, batch_size=[3, 4]) >>> td0, td1, td2 = td.unbind(0) >>> td0['x'] tensor([0, 1, 2, 3]) >>> td1['x'] tensor([4, 5, 6, 7])
Every tensordict class and every tensorclass has this method. For the other methods, see Methods by task.
