speed up evaluation #20
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
since we set mr=4 as default while training, I do the same thing to evaluation, so the evaluation is in mr=4 level which saves cuda memory and speed up.
Also the curve of mask mAP is now closer to cocoapi. Check https://wandb.ai/laughing/evaluation-test?workspace=user-laughing.
baseis the old one, 24.1(dashboard) vs 26.7(cocoapi), which got a big jump.optimizedis the speed up version, 26.5(dashboard) vs 26.9(cocoapi).In my case, the speed tasks 1min(3.5mins before) and pycocotool evaluation tasks 10mins(18mins before) now.
Btw since I have different coco path, I got the cocoapi result manually which the dashboard didn't show it.