Skip to content

Fork adapted to Mac users, now available #24

@fnachon

Description

@fnachon

Hello,

Proteina-Complexa is really great, and it would be useful to adapt it to the Apple Silicon Platform for testing by people who do not have access to top NVIDIA hardware, such as students.

For information, I succeeded in creating a working fork for Apple Silicon (MPS/Metal). It is currently HERE.

It runs in a Python 3.13 conda environment with PyTorch 2.11, jax-mps, and RF3 and tmol forks that I also adapted to Apple Silicon (thanks to Codex and Claude code!).

I can make a PR, though I understand it is not of interest to NVIDIA.

Florian

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions