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Outdated suggestions cannot be applied.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
[Torch] Fix PyTorch NMS conversion for negative scores #7137
[Torch] Fix PyTorch NMS conversion for negative scores #7137
Changes from all commits
5cf4ee4
cf401cf
5966041
9af6eda
63c43fb
39e007d
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.
Glad to see
rpn_pre_nms_top_n_test
is able to limit the proposals before nms. I am not sure if the parameter is specified for real use cases, seems using default of the parameter to do benchamarking makes more sense to meThere 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 think the default parameter 1000 they picked is fairly conservative. This means for each level in the feature pyramid, of which there is 5 if we use resnet 50 backbone, we get maximum of 1000 x 5 boxes as input to RPN. They have another parameter
rpn_post_nms_top_n_test
, which is like topk applied after NMS. This value is also by default 1000 and it is not per class unlikerpn_pre_nms_top_n_test
. This means we always have 1000 boxes after NMS regardless ofrpn_pre_nms_top_n_test
.