torch.Tensor.to_dense — PyTorch 2.7 documentation (original) (raw)

Tensor.to_dense(dtype=None, *, masked_grad=True) → Tensor¶

Creates a strided copy of self if self is not a strided tensor, otherwise returns self.

Keyword Arguments

• {dtype} –
• masked_grad (bool, optional) – If set to True (default) andself has a sparse layout then the backward ofto_dense() returns grad.sparse_mask(self).

Example:

s = torch.sparse_coo_tensor( ... torch.tensor([[1, 1], ... [0, 2]]), ... torch.tensor([9, 10]), ... size=(3, 3)) s.to_dense() tensor([[ 0, 0, 0], [ 9, 0, 10], [ 0, 0, 0]])