v1.7.2

What's Changed

• Fix type error in faster_eval_api by @matteomrz in #65
• update workflow by @MiXaiLL76 in #66
• Chore GitHub workflows: refine triggers, align naming, and improve matrices by @Borda in #67
• Fix indentation in unittest.yml for branch specification by @Borda in #68
• Enhanced README with detailed benefits, installation options, performance benchmarks, and feature highlights by @Borda in #69
• Add extensive pycocotools comparison tests with large synthetic datasets by @Copilot in #71
• Fix: extended_metrics raises clear ValueError when IoU=0.50 absent from iouThrs by @Copilot in #75
• Fix extended_metrics precision over-estimation due to monotone PR-curve interpolation by @Copilot in #74
• update version by @MiXaiLL76 in #76

New Contributors

• @matteomrz made their first contribution in #65
• @Copilot made their first contribution in #71

Full Changelog: 1.7.0...1.7.2

v1.7.0

What's Changed

• Improve visual by @MiXaiLL76 in #60
• add nano and micro ranges for very small objects by @jean-louquetin in #61
• The path to solving memory problems #57 by @MiXaiLL76 in #62

New Contributors

• @jean-louquetin made their first contribution in #61

Full Changelog: 1.6.7...1.7.0

v1.6.7

v1.6.7

• Added workflow for automatic Sphinx documentation build and deploy (.github/workflows/docs.yml)
• Improved main build workflow (build.yml): now includes documentation generation task and enhanced handling of environment variables and package publishing
• Updated testing workflow (test.yaml): added support for multiple Python versions and triggering via workflow_call
• Major refactoring and optimization of the C++ core (cocoeval.cpp, cocoeval.h, new dataset.cpp): improved readability, redesigned data structures, and optimized evaluation logic for better speed
• Significant improvements to annotation parsing, serialization, and evaluation handling in the C++ part
• Added new hooks and formatters (e.g., clang-format) to .pre-commit-config.yaml, cleaned up and structured existing sections
• Updated README.md: added comparison table between pycocotools and faster-coco-eval, improved project advantages section
• Performed code decomposition, optimization, and standardization of variables in workflow files
• Modern C++ idioms applied throughout the codebase for better readability and performance

What's Changed

• Std20 by @MiXaiLL76 in #54
• update docs by @MiXaiLL76 in #55

Full Changelog: 1.6.6...1.6.7

v1.6.6

v1.6.6

• Support pytorch dist FasterCocoEvaluator
• Support torchvision CocoDetection Dataset as FasterCocoDetection
• Return COCO.showAnns #52

v1.6.5

• Support CrowdPose Dataset
  • add CrowdPose simple test
• ImageEvaluation pickleable #45
• Create test for MP eval https://github.com/MiXaiLL76/faster_coco_eval/blob/main/tests/test_world_evalutor.py

v1.6.4

• fix windows mask api.
• numpy>=2 support for py39+
• update CI\CD for build macos & windows whl

1.6.4

v1.6.4

• fix windows mask api.
• numpy>=2 support for py39+
• update CI\CD for build macos & windows whl

Full Changelog: 1.6.3...1.6.4

v1.6.3

What's Changed

• fix windows build by @MiXaiLL76 in #40
• Fix torchmetric bug by @MiXaiLL76 in #41
• Macos build fix by @MiXaiLL76 in #42
• convert linting to pre-commit & Ruff by @Borda in #43

New Contributors

• @Borda made their first contribution in #43

Full Changelog: 1.6.0...1.6.3

1.6.0

v1.6.0

• Rework mask_api with pybind11 C++ .
• Rework RLE support.
• Create test files for all components.
• The math_matches function has been reworked, with an emphasis on using C++ code.
• Added more documentation of functions. Optimized existing ones.
• Added rleToBoundary func with 2 backend ["mask_api", "opencv"]
• IoU type boundary support (further testing is needed)
• Create async rle and boundary comput discussion

1.5.7

What's Changed

• Fix get ann ids by @MiXaiLL76 in #36

Full Changelog: 1.5.6...1.5.7

1.5.6

v1.5.6

• Replace CED MSE curve with MAE (px) curve
• Add CED examples
• Display IoU and MAE for keypoints
• Reworked eval._prepare to clear up the return flow
• Reworked the C++ part of COCOevalEvaluateImages and COCOevalAccumulate
  • Add new COCOevalEvaluateAccumulate to combine these two calls. You can use old style separate_eval==True (default=False)
  • COCOevalAccumulate & COCOevalEvaluateAccumulate -> COCOeval_faster.eval is now correctly created as numpy arrays.
• Append LVIS dataset support lvis_style=True in COCOeval_faster

cocoEval = COCOeval_faster(cocoGt, cocoDt, iouType, lvis_style=True, print_function=print)
cocoEval.params.maxDets = [300]

New Contributors

• @JohannesTheo made their first contribution in #32

Full Changelog: 1.5.5...1.5.6

1.5.5

What's Changed

• v1.5.5 by @MiXaiLL76 in #30

Full Changelog: 1.5.4...1.5.5

v1.5.5

• Add CED MSE curve
• Review tests
• Review COCOeval_faster.math_matches function and COCOeval_faster.compute_mIoU function
• Add img+category to ann ids mapping via COCO.img_cat_ann_idx_map
• Add img to ann ids mapping via COCO.img_ann_idx_map