The included Dockerfile uses a pytorch:2.3.0-cuda12.1-cudnn8-runtime base image and includes logic for the installation of the Genie 2 requirements. It also pulls the base checkpoint from git lfs (though it appears that the aqlboratory account is over its LFS quota).
The goal of this image is to make it easier for users to use Genie 2 without needing to install anything, which is perfect for scaling sequence/structure generation in high-performance computing environments.
The included Dockerfile uses a
pytorch:2.3.0-cuda12.1-cudnn8-runtimebase image and includes logic for the installation of the Genie 2 requirements. It also pulls the base checkpoint fromgit lfs(though it appears that the aqlboratory account is over its LFS quota).The goal of this image is to make it easier for users to use Genie 2 without needing to install anything, which is perfect for scaling sequence/structure generation in high-performance computing environments.