Allow tracking on Instances (by adding Instance.tracking_score)

[roomrys]

Description

Prior to Allow Retracking, we did not allow any tracking on user instances. Since re-tracking enables users to run tracking on user instances, we now need to either add a tracking_score attribute to user-instances, or refrain from assigning the tracking score if the matched instance is a user instance (see #1301). This PR adds an Instance.tracking_score attribute and updates spots where the tracking_score is written (including a FORMAT_ID bump in LabelsV1Adaptor).

Types of changes

• Bugfix (Should have done this in Allow retracking #867)
• New feature (Allow tracking on instances)
• Refactor / Code style update (no logical changes)
• Build / CI changes
• Documentation Update
• Other (explain)

Does this address any currently open issues?

• Allow Tracking-Only Pipeline (GUI & CLI) #1296

Outside contributors checklist

• Review the guidelines for contributing to this repository
• Read and sign the CLA and add yourself to the authors list
• Make sure you are making a pull request against the develop branch (not main). Also you should start your branch off develop
• Add tests that prove your fix is effective or that your feature works
• Add necessary documentation (if appropriate)

Thank you for contributing to SLEAP!

❤️

[codecov]

Codecov Report

Merging #1302 (1ea047d) into develop (b5aab0e) will decrease coverage by 0.01%.
The diff coverage is 86.36%.

@@             Coverage Diff             @@
##           develop    #1302      +/-   ##
===========================================
- Coverage    72.36%   72.36%   -0.01%     
===========================================
  Files          133      133              
  Lines        23662    23675      +13     
===========================================
+ Hits         17123    17132       +9     
- Misses        6539     6543       +4     

Impacted Files	Coverage Δ
sleap/nn/tracking.py	72.87% <62.50%> (-0.23%)	⬇️
sleap/gui/app.py	74.80% <100.00%> (ø)
sleap/gui/widgets/docks.py	94.69% <100.00%> (+0.37%)	⬆️
sleap/info/write_tracking_h5.py	90.90% <100.00%> (ø)
sleap/instance.py	89.07% <100.00%> (ø)
sleap/io/format/hdf5.py	94.17% <100.00%> (-0.03%)	⬇️
sleap/io/format/nix.py	95.53% <100.00%> (-0.02%)	⬇️

... and 7 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[talmo]

Quick question: what happens when there is a user Instance that was created from a PredictedInstance on a frame? Will the user Instance be used preferentially and the PredictedInstance ignored? If not, this may lead to issues where extraneous tracks are generated right?

[roomrys]

All instances in the labeled frame were being considered individually. I've added the following to prefer user labeled instances in the frame if any:

sleap/sleap/nn/tracking.py

Lines 1192 to 1213 in 1ea047d

	# Prefer user instances over predicted instances
	instances = []
	if lf.has_user_instances:
	instances_to_track = lf.user_instances
	if lf.has_predicted_instances:
	instances = lf.predicted_instances
	else:
	instances_to_track = lf.predicted_instances
	track_args = {"untracked_instances": instances_to_track}
	if tracker.uses_image:
	track_args["img"] = lf.video[lf.frame_idx]
	else:
	track_args["img"] = None
	instances.extend(tracker.track(**track_args))
	new_lf = LabeledFrame(
	frame_idx=lf.frame_idx,
	video=lf.video,
	instances=instances,
	)

[roomrys]

Holding off on this PR since it would require a file format update... best to be worn in on the develop branch for a bit before tacking onto the release. This mean that retracking can only be done on prediction-only projects until this PR is merged (i.e. users need to remove their user-labeled instances before retracking to avoid error).