Fix and relax bboxes requirements

[sfmig]

Description

What is this PR

• Bug fix
• Addition of a new feature
• Other

Why is this PR needed?
The current implementation gives an error when loading a dataset with a single individual.

It also has some inconveniences that I realised were a bit too strict when I started using it more. Mainly:

• the requirement for the encoding of the frames number in the filename - right now we expect the number to be prefixed by frame_, but often the filename is simply the frame number, so I removed this.
• when loading the frames as specified in the input file, it may be that the frames are not regularly spaced.
  • Currently this would throw an error because we check the frames are separated by exactly 1.
  • I relaxed this requirement to simply require the frame numbers increase.
  • If we extract them from the VIA file they are sorted before creating the dataset, but if the frame array is passed using from_numpy it is up to the user to sort it.

What does this PR do?

• Adds fix to load a bounding boxes' dataset of a single animal
  • fixes issue with squeezing
  • extends existing tests to a file with a single animal
• Relaxes the requirement for extracting the frame number from the filename
  • before we required a frame_ prefix
• Relaxes the requirement on the loaded frames being continuous
  • continuous: frames increase exactly by one number
  • instead we just check the frame numbers always increase
  • two fixtures are added to test_datasets_validators/test_bboxes_dataset_validator_frame_array to check this behaves as expected.
• Simplifies the fixtures for the load_bboxes tests
  • Uses fixture "factories" to reduce the number of predefined fixtures
  • Adapts existing tests to use those fixtures

References

Many of these issues were identified while working on #312

How has this PR been tested?

Tests pass locally and on CI.

Is this a breaking change?

No.

Does this PR require an update to the documentation?

The relevant docstrings have been updated.

Checklist:

• The code has been tested locally
• Tests have been added to cover all new functionality
• The documentation has been updated to reflect any changes
• The code has been formatted with pre-commit

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 99.78%. Comparing base (4de4963) to head (489cba1).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #313   +/-   ##
=======================================
  Coverage   99.78%   99.78%           
=======================================
  Files          14       14           
  Lines         912      927   +15     
=======================================
+ Hits          910      925   +15     
  Misses          2        2           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[sonarqubecloud]

Quality Gate passed

Issues
7 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[niksirbi]

Looks good @sfmig! That's some fixture factory extravaganza right there!

I only left very few clarifying comments/questions.

I also think you may have forgotten to update the docstring of the from_numpy() functions. It currently says:

frame_array : np.ndarray, optional
        Array of shape (n_frames, 1) containing the frame numbers for which
        bounding boxes are defined. If None (default), frame numbers will
        be assigned based on the first dimension of the ``position_array``,
        starting from 0. If a specific array of frame numbers is provided,
        these need to be consecutive integers.

I think they no longer need to be consecutive, right?

[niksirbi]

Just a few more comments

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud