google-deepmind / google-deepmind/alphafold
Based on equation 46 in Supplementary Info, should the clash loss be divided by number of atoms?
Open
- Dominant language
- Python
- Stars
- 14.9k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
As the subject line indicates, I was confused by the division of structural clash loss by the number of atoms. The equation in the paper indicates a overall summation, not an average.
I'd appreciate any clarification.
https://github.com/google-deepmind/alphafold/blob/2ca41b1991bbf17be4a86c1face55597ba4c7621/alphafold/model/folding.py#L731
Contributor guide
Assessment
This issue has not been assessed yet.