The following methods are specific to sparse CSR tensors: Returns True if self is a sparse COO tensor that is coalesced, False otherwise. Removes all specified elements from a sparse tensor self and resizes self to the desired size and the number of sparse and dense dimensions. Resizes self sparse tensor to the desired size and the number of sparse and dense dimensions. Returns a coalesced copy of self if self is an uncoalesced tensor. The following Tensor methods are specific to sparse COO tensors: Return the values tensor of a sparse COO tensor. Return the indices tensor of a sparse COO tensor. Returns a new sparse tensor with values from a strided tensor self filtered by the indices of the sparse tensor mask.Ĭonvert a tensor to compressed row storage format. Return the number of sparse dimensions in a sparse tensor self. Return the number of dense dimensions in a sparse tensor self. Is True if the Tensor uses sparse storage layout, False otherwise. The following Tensor methods are related to sparse tensors: All PyTorch operations,Įxcept torch.smm(), support backward with respect to strided Where “Sparse grad?” column indicates if the PyTorch operation supportsīackward with respect to sparse matrix argument. Multiplication, and is matrix multiplication. Scalar (float or 0-D PyTorch tensor), * is element-wise M denotes a matrix (2-D PyTorch tensor), and Vĭenotes a vector (1-D PyTorch tensor). Sparse matrices where the operands layouts may vary. The following table summarizes supported Linear Algebra operations on CPU threading and TorchScript inference. CUDA Automatic Mixed Precision examples.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |