Evaldet

Fixed

* Fix typing for compatibility with Python 3.8 ([22](https://github.com/tadejsv/EvalDeT/pull/22))

Changed

* Fix publishing with `hatch` ([21](https://github.com/tadejsv/EvalDeT/pull/21))

Changed

* Fix correct 1-index frame handling for MOT-based files ([20](https://github.com/tadejsv/EvalDeT/pull/20))
* Fix handling of "outside" detections in CVAT format ([20](https://github.com/tadejsv/EvalDeT/pull/20))

* Switch to `hatch` for packaging ([20](https://github.com/tadejsv/EvalDeT/pull/20))

Changed
* Performance improvement for HOTA metrics using sparse matrices ([19](https://github.com/tadejsv/EvalDeT/pull/19))

Changed

* MOT Metrics are now computed using the `MOTMetrics` class, which enables efficient sharing of pre-computed IoU distances across metrics ([17](https://github.com/tadejsv/EvalDeT/pull/12))
* Increase minimum Python version to 3.9 ([16](https://github.com/tadejsv/EvalDeT/pull/16))
* Switch packaging system to Poetry ([16](https://github.com/tadejsv/EvalDeT/pull/16))

Added

* More integration tests (though for some reason, there are discrepancies with official results at times) ([12](https://github.com/tadejsv/EvalDeT/pull/12))
* Filtering method for tracks (`filter_by_class`, `filter_by_conf`) ([12](https://github.com/tadejsv/EvalDeT/pull/12))
* General method for reading CSV file (`from_csv`), reading MOT gt files (`from_mot_gt`) ([12](https://github.com/tadejsv/EvalDeT/pull/12))


Changed

* Version large files for integration tests with `git-lfs` ([12](https://github.com/tadejsv/EvalDeT/pull/12))
* All `Track` class methods for reading CSV files become just a wrapper around `from_csv` ([12](https://github.com/tadejsv/EvalDeT/pull/12))