torch.Tensor.narrow_copy¶
-
Tensor.
narrow_copy
(dimension, start, length) → Tensor¶ Same as
Tensor.narrow()
except returning a copy rather than shared storage. This is primarily for sparse tensors, which do not have a shared-storage narrow method. Callingnarrow_copy
withdimemsion > self.sparse_dim()
will return a copy with the relevant dense dimension narrowed, andself.shape
updated accordingly.