Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
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 Reverse Distillation #343
✨ Add Reverse Distillation #343
Changes from 1 commit
0b6d836
7dd75b1
63d2921
253c215
9d86107
6ae8a8b
8e4e060
83494a2
6b63541
73c2851
8171956
b4a2369
0520a07
d187625
5114b9b
6d1102a
f58e6ec
6f96e90
1c71642
cf377bd
6a4c66c
149e3c0
94ad970
File filter
Filter by extension
Conversations
Jump to
There are no files selected for viewing
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.
Are the benchmarking results still the same after this change?
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.
I am running it right now. But that was a good catch. I saw that the encoder was in training mode in the train step.
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.
Interestingly apart from pixel level score, image level scores are lower
![image](https://user-images.githubusercontent.com/17232914/172832983-374f1f15-66d1-4081-b93b-71eec6f27bc8.png)
![image](https://user-images.githubusercontent.com/17232914/172833009-d0d7eb98-ed75-4685-82cf-0846254522c6.png)
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.
The average results also seem to be a bit lower than those reported in the original paper (98.5% image AUROC, 97.8% pixel AUROC). For now I would suggest to merge this PR, but it would be good to investigate if this difference can be explained.
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.
Does the official implementation give the exact same results reported in the paper?