New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add MinkowskiEngine #23
Conversation
@akisman Thanks for following the style of the previous PR for this add. Can you do the following:
If that all looks good, we'll be able to merge. |
@ejm714 Sure thing! I've pasted the image size for CPU and GPU below (Github actions build both without issue) and the log file form Image Size
Example Submission Log
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks @akisman!
Noting for posterity that this PR does increase each image by around 5GB. This matters when people pull this image to test locally. Given the general large size of the data in this competition, we'll assume this is workable.
This pull request adds MinkowskiEngine, an auto-diff neural network library for high-dimensional sparse tensors from NVIDIA.
Because the base CUDA image doesn't include the development tools needed to build the library I've updated it to the devel image. Hopefully this isn't an issue since it's in essence the same image but also includes headers & development tools. Reference: https://hub.docker.com/r/nvidia/cuda