pytorch_sparse
pytorch_sparse copied to clipboard
rusty1s
PyTorch Extension Library of Optimized Autograd Sparse Matrix Operations
I attempted to compile pytorch_sparse version 0.6.18 using C++ with the commands provided on the homepage, but I encountered an error. 4 warnings generated. [ 54%] Building CXX object CMakeFiles/torchsparse.dir/csrc/diag.cpp.o...
Exploring the behavior of pytorch geometric's `LinkNeighborSampler`, we identified the sampling method consistently oversampled the first possible combination of neighbors, while consistently missed the last possible combination of neighbors. This...
Before this fix uniform sampling is not guaranteed, over-representing the first combination of neighbors, and not sampling the last possible combination. This PR ensures all neighbor combinations are considered during...
I installed metis and then manually installed torch_sparse, but it showed RuntimeError: Not compiled with METIS support
The problem occured when I try to pass a SparseTensor to PyG GCNconv. I'm working with python 3.10, cuda 12.1, torch 2.2.0, PyG 2.5.2 and torch_sparse 0.6.18 installed by conda...
Hi, I am using pytorch_sparse matmul, but see abnormal result from torch_sparse import SparseTensor, matmul # x [batch, features, nodes, step] x = torch.rand(100, 20, 3, 24) adj = torch.eye(100,100)...
torch.compile fails for SparseTensor.matmul torch version: 2.6.0 pytorch_sparse version: 0.6.18 py312hf276b08_7 from conda-forge reproducer: ``` import torch from torch_sparse import SparseTensor import scipy.sparse as sp import numpy as np def...
Hello everyone, I have the following issue using torch-sparse: CUDA Version: 12.4 Architecture: aarch64 OS: Ubuntu 22.04.5 LTS This is what I did: `conda create -n test python=3.11` `conda activate...
The weight2metis function may encounter issues when min_diff == 0, which can result from consecutive identical values in the input tensor. Is it the expected behavior?
