search

jasonkyuyim / se3_diffusion

Implementation for SE(3) diffusion model with application to protein backbone generation
https://arxiv.org/abs/2302.02277
MIT License
305 stars 50 forks source link

Clarification on auxiliary loss weight #21

Closed amorehead closed 1 year ago

amorehead commented 1 year ago

Hello. Is the following aux_loss_weight supposed to be 0.25 or instead 25 as mentioned in the arXiv manuscript (i.e., for loss weight w)?

https://github.com/jasonkyuyim/se3_diffusion/blob/8c7a328fe9297f81b72694518aebb010a1f46169/config/base.yaml#L110

jasonkyuyim commented 1 year ago

Nice catch! It should be 0.25. I'll be updating the arxiv with the ICML accepted version in the coming weeks with this fix.

Thanks a lot!