Shortcuts

torch.Tensor.to_sparse_csr

Tensor.to_sparse_csr(dense_dim=None) Tensor

Convert a tensor to compressed row storage format (CSR). Except for strided tensors, only works with 2D tensors. If the self is strided, then the number of dense dimensions could be specified, and a hybrid CSR tensor will be created, with dense_dim dense dimensions and self.dim() - 2 - dense_dim batch dimension.

Parameters:

dense_dim (int, optional) – Number of dense dimensions of the resulting CSR tensor. This argument should be used only if self is a strided tensor, and must be a value between 0 and dimension of self tensor minus two.

Example:

>>> dense = torch.randn(5, 5)
>>> sparse = dense.to_sparse_csr()
>>> sparse._nnz()
25

>>> dense = torch.zeros(3, 3, 1, 1)
>>> dense[0, 0] = dense[1, 2] = dense[2, 1] = 1
>>> dense.to_sparse_csr(dense_dim=2)
tensor(crow_indices=tensor([0, 1, 2, 3]),
       col_indices=tensor([0, 2, 1]),
       values=tensor([[[1.]],

                      [[1.]],

                      [[1.]]]), size=(3, 3, 1, 1), nnz=3,
       layout=torch.sparse_csr)

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources