torch.Tensor.to_sparse#

Tensor.to_sparse(sparseDims)→Tensor#

Returns a sparse copy of the tensor. PyTorch supports sparse tensors incoordinate format.

Parameters

sparseDims (int,optional) – the number of sparse dimensions to include in the new sparse tensor

Example:

>>>d=torch.tensor([[0,0,0],[9,0,10],[0,0,0]])>>>dtensor([[ 0,  0,  0],        [ 9,  0, 10],        [ 0,  0,  0]])>>>d.to_sparse()tensor(indices=tensor([[1, 1],                       [0, 2]]),       values=tensor([ 9, 10]),       size=(3, 3), nnz=2, layout=torch.sparse_coo)>>>d.to_sparse(1)tensor(indices=tensor([[1]]),       values=tensor([[ 9,  0, 10]]),       size=(3, 3), nnz=1, layout=torch.sparse_coo)

to_sparse(*,layout=None,blocksize=None,dense_dim=None)→Tensor

Returns a sparse tensor with the specified layout and blocksize. Iftheself is strided, the number of dense dimensions could bespecified, and a hybrid sparse tensor will be created, withdense_dim dense dimensions andself.dim() - 2 - dense_dim batchdimension.

Note

If theself layout and blocksize parameters matchwith the specified layout and blocksize, returnself. Otherwise, return a sparse tensor copy ofself.

Parameters

• layout (torch.layout, optional) – The desired sparselayout. One oftorch.sparse_coo,torch.sparse_csr,torch.sparse_csc,torch.sparse_bsr, ortorch.sparse_bsc. Default: ifNone,torch.sparse_coo.

• blocksize (list, tuple,torch.Size, optional) – Block sizeof the resulting BSR or BSC tensor. For other layouts,specifying the block size that is notNone will result in aRuntimeError exception. A block size must be a tuple of lengthtwo such that its items evenly divide the two sparse dimensions.

• dense_dim (int,optional) – Number of dense dimensions of theresulting CSR, CSC, BSR or BSC tensor. This argument should beused only ifself is a strided tensor, and must be avalue between 0 and dimension ofself tensor minus two.

Example:

>>>x=torch.tensor([[1,0],[0,0],[2,3]])>>>x.to_sparse(layout=torch.sparse_coo)tensor(indices=tensor([[0, 2, 2],                       [0, 0, 1]]),       values=tensor([1, 2, 3]),       size=(3, 2), nnz=3, layout=torch.sparse_coo)>>>x.to_sparse(layout=torch.sparse_bsr,blocksize=(1,2))tensor(crow_indices=tensor([0, 1, 1, 2]),       col_indices=tensor([0, 0]),       values=tensor([[[1, 0]],                      [[2, 3]]]), size=(3, 2), nnz=2, layout=torch.sparse_bsr)>>>x.to_sparse(layout=torch.sparse_bsr,blocksize=(2,1))RuntimeError: Tensor size(-2) 3 needs to be divisible by blocksize[0] 2>>>x.to_sparse(layout=torch.sparse_csr,blocksize=(3,1))RuntimeError: to_sparse for Strided to SparseCsr conversion does not use specified blocksize>>>x=torch.tensor([[[1],[0]],[[0],[0]],[[2],[3]]])>>>x.to_sparse(layout=torch.sparse_csr,dense_dim=1)tensor(crow_indices=tensor([0, 1, 1, 3]),       col_indices=tensor([0, 0, 1]),       values=tensor([[1],                      [2],                      [3]]), size=(3, 2, 1), nnz=3, layout=torch.sparse_csr)