Skip to content

Version of pytorch-metis that has bindings that directly accept pytorch tensors for constructing metis graphs. No data copies are required to convert from pytorch CSR and Metis's graph representation because METIS uses CSR internally.

License

Notifications You must be signed in to change notification settings

MITIBMxGraph/torch-metis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

30 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

METIS for Python (with support for PyTorch CSR tensors)

The following is a version of the "METIS wrapper using ctypes" that was originally written by Ken Watford ([email protected]). The original contents of the README file from his repository (https://github.com/kw/metis-python) are included at the end of this document. All modifications are made under this software's original MIT license, and no guarantee is provided as to their correctness.

This package is modified to support PyTorch tensors as input to make it more convienient to partition graphs in PyTorch. Presently, this repository is primarily intended to be used in conjunction with the software for SALIENT++ for performing graph partitioning.

Original README : METIS for Python

Wrapper for the METIS library for partitioning graphs (and other stuff).

This library is unrelated to PyMetis, except that they wrap the same library. PyMetis is a Boost Python extension, while this library is pure python and will run under PyPy and interpreters with similarly compatible ctypes libraries.

NetworkX is recommended for representing graphs for use with this wrapper, but it isn't required. Simple adjacency lists are supported as well.

Please see the full documentation for examples or the GitHub repository for bug reports

About

Version of pytorch-metis that has bindings that directly accept pytorch tensors for constructing metis graphs. No data copies are required to convert from pytorch CSR and Metis's graph representation because METIS uses CSR internally.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages