You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The classifier should include some summary about the novelty of a track in the JSON output.
It's not clear how to properly summerise the novelty of a track, but here are some ideas
Average.
Max / 90th percentile.
Minimum / 10th percentile.
Novelty at time of highest confidence
These will all tell us slightly different things. My guess is that 10th percentile or novelty at time of highest confidence is the way to go, as we don't really care if the input was novel during unimportant parts of the tracking.
Potentially all 4 could be exported.
The text was updated successfully, but these errors were encountered:
The classifier should include some summary about the novelty of a track in the JSON output.
It's not clear how to properly summerise the novelty of a track, but here are some ideas
Average.
Max / 90th percentile.
Minimum / 10th percentile.
Novelty at time of highest confidence
These will all tell us slightly different things. My guess is that 10th percentile or novelty at time of highest confidence is the way to go, as we don't really care if the input was novel during unimportant parts of the tracking.
Potentially all 4 could be exported.
The text was updated successfully, but these errors were encountered: