Edge cases for empty frames + test with non-empty frames.#1395
Merged
Conversation
Collaborator
|
Think you're missing the logic fix here - I only see tests |
Collaborator
Author
|
That was really sad. Pushed. |
This file contains hidden or 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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In the BOEM workflow I found a funny edge case where if the test dataset has both empty frames and non-empty frames, it produces a non-intuitive results. It was mathematically correct, but just confusing and too literal. We were dividing the recalls by the total number of images, so the average box_recall score was (number of matches/total number of images), but given that some of those images are empty, it produces a strange result. So if you had two images, 1 with one ground truth, correctly matched, and one that was empty, the average box_recall was 0.5, when it really should be 1 match /1 image with GT + empty image, not 1 match / (2 images).
AI-Assisted Development
I have coalesced around a cursor strategy where I identify errors, discuss with the agent. I then have cursor develop failing tests, which then I implement a correction for and then we all agree its passing. This feels reasonable since there is an adversarial element to it.