Skip to content

Conversation

tklijnsma
Copy link
Contributor

@tklijnsma tklijnsma commented Oct 12, 2023

This PR contains a C++ extension for object condensation. The previous pure-Python version suffered from poor memory optimization - that problem should now be resolved.

Currently, the extension runs on CPU only. Writing a CUDA kernel based on the C++ CPU template should however not be too hard, and will be done in a future PR.

See tests/test_oc.py for an example of how to run the extension.

@jkiesele
Copy link
Contributor

We have helper ops in TF for that purpose, you could just port them. You can find them in oc_helpers.py

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants