3 Commits

Author	SHA1	Message	Date
PyTorch MergeBot	43ae3073f9	Revert "[traced-graph][sparse] remove redundant assert in sparse prop test (#128523)" ... This reverts commit ba3726d02b. Reverted https://github.com/pytorch/pytorch/pull/128523 on behalf of https://github.com/DanilBaibak due to Sorry for the revert. Looks like your changes broke the inductor tests: inux-jammy-cpu-py3.8-gcc11-inductor, linux-jammy-cpu-py3.8-gcc11-inductor, linux-jammy-cpu-py3.8-gcc11-inductor. [Here you can find more details](ba3726d02b). ([comment](https://github.com/pytorch/pytorch/pull/128523#issuecomment-2167518145))	2024-06-14 08:27:05 +00:00
Aart Bik	ba3726d02b	[traced-graph][sparse] remove redundant assert in sparse prop test (#128523) ... The assertEqualMeta() method already tests that the first argument is a FakeTensor https://github.com/pytorch/pytorch/issues/117188 Pull Request resolved: https://github.com/pytorch/pytorch/pull/128523 Approved by: https://github.com/soulitzer	2024-06-14 02:34:51 +00:00
Aart Bik	ff82e2e7cf	[traced-graph][sparse] propagate sparsity metadata into traced graph (#117907) ... Propagate sparsity metadata from sparse tensors of torch.sparse into the traced graph representation (with would be useful for a JIT backend that supports a "sparse compiler"). This is a first careful attempt, since the actual "meta" feature seem still incomplete for coo and completely lacking for csr/csc/bsr/bsc. For background see forum postings (with examples): https://discuss.pytorch.org/t/connecting-pytorch-sparse-tensors-with-mlir/195145 https://dev-discuss.pytorch.org/t/connecting-pytorch-sparse-tensors-with-mlir/1803 And feature request: https://github.com/pytorch/pytorch/issues/117188 Pull Request resolved: https://github.com/pytorch/pytorch/pull/117907 Approved by: https://github.com/pearu, https://github.com/ezyang	2024-05-23 22:46:46 +00:00