Skip segmentations of already processed subjects

What's Changed

• Avoid re-segmenting already processed subjects by @clementpoiret in #28
• refactor: refactor unnecessary else / elif when if block has a return statement by @deepsource-autofix in #29
• refactor: remove commented out code by @deepsource-autofix in #30
• refactor: remove assert statement from non-test files by @deepsource-autofix in #32
• style: format code with isort and Yapf by @deepsource-autofix in #31
• refactor: put docstring into a single line by @deepsource-autofix in #36
• refactor: remove blank lines after docstring by @deepsource-autofix in #35
• refactor: merge collapsible if statements by @deepsource-autofix in #34
• refactor: use identity check for comparison to a singleton by @deepsource-autofix in #33

New Contributors

• @deepsource-autofix made their first contribution in #29

Full Changelog: v1.2.0...v1.2.1

v1.2.0 - More data, better architecture, hosted on HuggingFace

What's Changed

• Fix hydra warnings by @ylep in #23
• fix: remove irrelevant warning by @ylep in #24
• fix: use distinct directory for bagging_sq models by @ylep in #25
• From Zenodo to HuggingFace, updated models, and finetuning scripts by @clementpoiret in #26

New Contributors

• @ylep made their first contribution in #23

Full Changelog: v1.1.3...v1.2.0

v1.1.3 - Lower onnxruntime min version for compatibility

Full Changelog: v1.1.2...v1.1.3

v1.1.2 - Fix macOS installation and update deps

Full Changelog: v1.1.1...v1.1.2

v1.1.1 - Whole hippocampus segmentation

Quick update to add whole hippocampus segmentation

v1.1.0 - Multispectral mode, Sparsity and int8-Quantization

What's Changed

• New optional multispectral mode de segment from both T1 AND T2 images by @clementpoiret in #20 and #19
• Bug fixes and optimization.
• Models trained with more data,
• Sparse and int8-Quantized models.

Full Changelog: v1.0.1...v1.1.0

v1.0.1 - Bug fixes and Optimizations

What's Changed

• Hotfix constant batch_size by @clementpoiret in #14
• Memory Optimizations

Full Changelog: v1.0.0...v1.0.1

v1.0.0 - Stable Release

This is the first stable release of HSF 🔥

What's Changed

HSF

• Added aleatoric uncertainty maps for each prediction to analyze segmentation results,
• Support for NeuralMagic's DeepSparse for fast inference on AVX512(VNNI)-capable CPUs,
• Support for batch processing through the argument hardware.engine_settings.batch_size (defaults to 1),
• Checking for updates at startup,
• Lowered min python version to 3.7,
• Bug fixes and optimizations.

Models

• Trained with more T1w and T2w MRIs,
• Trained on more hippocampal sclerosis and Alzheimer's disease cases,
• Updated training pipeline (hyperparameter tuning),
• single models are now independant from bags.

Full Changelog: v0.1.2...v1.0.0

Offsets and Execution Providers

Changelog:

• Added the possibility to add an offset to the initial bounding box detected by roiloc,
• Added the possibility to customize Hardware Execution Providers (defaults to "CUDAExecutionProvider" and "CPUExecutionProvider").

Hippocampal Segmentation Factory

Fixed CUDAExecutionProvider