rusty1s/pytorch_sparse
A PyTorch extension library providing optimized sparse matrix operations with automatic differentiation support.
Not currently ranked — collecting fresh signals.
star history
This package provides a collection of optimized sparse matrix operations with autograd support, including coalesce, transpose, sparse-dense matrix multiplication, and sparse-sparse matrix multiplication. All operations support both CPU and GPU execution and work with varying data types. The library simplifies sparse tensor manipulation by accepting index and value tensors directly rather than requiring torch.sparse_coo_tensor construction.
Frequently asked
- What is rusty1s/pytorch_sparse?
- A PyTorch extension library providing optimized sparse matrix operations with automatic differentiation support.
- Is pytorch_sparse open source?
- Yes — rusty1s/pytorch_sparse is open source, released under the MIT license.
- What language is pytorch_sparse written in?
- rusty1s/pytorch_sparse is primarily written in Python.
- How popular is pytorch_sparse?
- rusty1s/pytorch_sparse has 1.1k stars on GitHub.
- Where can I find pytorch_sparse?
- rusty1s/pytorch_sparse is on GitHub at https://github.com/rusty1s/pytorch_sparse.