Skip to content

factorisation subpaterns, float compatible, inplace and memory optimisation.... #10

Open
@robinechuca

Description

@robinechuca

Hi, I've also coded a function equivalent to this one on my side.
Thank you very much for publishing your project as open source.
One of the problems with open-source is its the diversification.
Our 2 projects seem very complementary, as they both have different strengths. Yours is more torch-friendly with elegant integration, mine is more optimized on the points described in the titles. I'd be happy to merge my project with yours. That way, I can delete mine. It's better to have one single library that does everything well than 2 that do everything halfway!

Here's the source code for my project, which is perhaps a little more "sympy" than "torch" oriented:
https://framagit.org/robinechuca/cutcutcodec/-/blob/main/cutcutcodec/core/compilation/sympy_to_torch.py

Are you interested to implement some of this optimizations in your project?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions