-
Notifications
You must be signed in to change notification settings - Fork 1.3k
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Refactor ava hook #567
Refactor ava hook #567
Conversation
Still WIP? Are there any configs need to be changed? Also give good warnings to help user migrate |
It would be great to add some migration notes in the top post, to give detailed info on what has been changed, and how to change configs/cmds to the new way. Example: open-mmlab/mmpose#382 |
Codecov Report
@@ Coverage Diff @@
## master #567 +/- ##
==========================================
- Coverage 84.76% 84.12% -0.64%
==========================================
Files 121 121
Lines 8579 8580 +1
Branches 1416 1416
==========================================
- Hits 7272 7218 -54
- Misses 953 1016 +63
+ Partials 354 346 -8
Flags with carried forward coverage won't be shown. Click here to find out more.
Continue to review full report at Codecov.
|
Do you need to update modelzoo, if there are related metrics affected? |
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.
Just changing names looks good to me. But how do we evaluate AR@AN and Recall?
Can metrics be ['mAP', 'AR@AN', 'Recall']
since those metrics have been calculated anyway?
fix #549
Migration note :
You need to use
--eval mAP
during AVA models testing now.AVADataset evaluation metrics modification:
Supported metrics before:
proposal
,bbox
Supported metrics after:
mAP
(the same asbbox
before), the oldproposal
is removed.