VCSparse: Generate fused and vectorizable kernels for sparse tensor programs on CPUs and GPUs. An web tool is available to try VCSparse online: https://tongzhou80.github.io/vcsparse-web/index.html.

More technical detals are to be added!

Installation

pip install vcsparse

Quick introduction

To compile a function with vcsparse, all you need to do is to add index notation to the function arguments, and decorate the function with @vcsparse.compile. Here's an example:

@vcsparse.compile(full_opt=True)
def mul_then_rowwise_sum(A: Tensor('i,j', 'csr'), B: Tensor('i,j', 'csr')):
    return (A * B).sum(1)

The option full_opt=True will enable all optimization passes. The directory tests/ contains many code examples of using VCSparse, feel free to check them out.

Note

Currently only a CPU backend (via Numba) is implemented. A GPU backend (via APPy) is under development.