jcjohnson / jcjohnson/densecap
Why gradients can not be propagated backward to the input proposal coordinates in the RoI pooling layer?
Open
- Dominant language
- Jupyter Notebook
- Stars
- 1.6k
- Forks
- 424
- PR merge metrics
- No merged PRs in 30d
Description
just like what i ask in title, in paper author said "gradients can be propagated backward from the output features to the input features, but not to the input proposal coordinates", i want to know why? thanks.
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.