pytorch_sparse
pytorch_sparse copied to clipboard
rusty1s
PyTorch Extension Library of Optimized Autograd Sparse Matrix Operations
ASSUME a Pytorch geometric layer - Layer(x, edge-index)
POC for spmm_sum: https://github.com/pytorch/pytorch/pull/83727 Links to other PRs: pyg - https://github.com/pyg-team/pytorch_geometric/pull/5498 pytorch_sparse - this pytorch - https://github.com/pytorch/pytorch/pull/85466
Hi, I find it when doing sparse-sparse multiplication with `SparseTensor` format, the speed is super slow even compared with doing this with dense format. The variable `s` and `adj_t` are...
Hi, I'm trying to install the latest version of torch-geometric. After trying multiple times the guide provided [here](https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html#), I'm still getting the below error while loading torch_sparse dependancy. ` Traceback...
Hi, I was wondering if you'd provide LU/Cholesky/QR factorization for sparse matrices? https://github.com/rgl-epfl/cholespy library provides for cholesky but one would need PD matrices. If I'd want to do LU, I...
Hi, there I have a tensor ten1 of shape [b,n,dim,i] which is sparse, and ten2 tensor of shape [i,m] and I want to output ten1@ten2 of shape [b,n,dim,m], how can...
Are there ROCm wheels for this project? https://github.com/Looong01/pyg-rocm-build has fairly older wheels, so was wondering if you expect to maintain/provide wheels for later ROCm versions.
Hi, I'm having the same problem with #174. I have two large adjacency matrices, the details are as follows adj_l SparseTensor(row=tensor([ 0, 0, 0, ..., 736388, 736388, 736388], device='cuda:2'), col=tensor([...
am using neighbour loader function in my code, which uses sample_adj_cpu function to sample neighbours. I am making some changes in this function which is present in the following file....
Running the benchmarks `main.py` targeting device `mps` results in an assertion failure: ``` Traceback (most recent call last): File "/Users/apullin/personal/pyg/pytorch_sparse/benchmark/main.py", line 174, in correctness(dataset) File "/Users/apullin/anaconda3/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context...
