v1.2.0#

We are pleased to announce the release of version 1.2.0! This update introduces valuable new features, performance improvements, bug fixes, and enhanced documentation to support a more stable and user-friendly experience.

This release has been tested on Python 3.10 to 3.13.

New Features#

  • Edge Visualisation: Introduced custom edge drawing for network visualisations to improve zooming and padding behavior.

  • Shape Combination: combine_shapes now fully supports a delete_existing parameter, enabling cleaner merges by optionally removing pre-existing shapes.

  • Boundary Estimation: Added automatic warnings and halts when estimate_boundary via alpha_shapes results in multiple shape outputs.

  • Boundary visualisation customisation : Added boundary_kwargs to visualise for enhanced control over boundary appearance in plots.

  • Simplify Shapes: Now optionally returns object IDs and allows specification of the target collection name for better downstream use.

Optimisations#

  • Parallel Proximity Computation: Distance computations for proximity networks with shape-objects have been parallelised for significant speed-ups.

  • Vectorised TDA Updates: Persistence updates have been vectorised to enhance performance in topological computations.

  • Alpha shape Efficiency: New alpha shape implementation vastly improving efficency by reducing the number of checks for finding shape boundaries and allowing for nested shapes.

Bug Fixes#

  • Label Inheritance Fix: Corrected an issue where labels were reset during object conversion—existing labels now retain their original color mappings.

  • QCM Tiling Deletion: Fixed a bug where mislabelled objects in the Quadrat Correlation Matrix prevented default tiling cleanup.

  • Cell Count Normalisation: Resolved issue where cell counts were incorrectly normalised by default.

  • Multiprocessing Stability: Wrapped network generation in a try-catch to avoid multiprocessing issues on UNIX systems.

  • wPCF index error: Addressed an index error in the weighted Pair Correlation Function (wPCF) when handling multiple populations in the same domain.

Documentation#

  • New Tutorials: Added guides on using distance and containment queries to the documentation.

  • Visualisation Defaults: Updated the default visualisation settings for network plots to reflect improved edge rendering behavior.

All tutorials and documentation have been updated to reflect the new features and changes in version 1.2.0.

We sincerely thank all contributors and users who provided feedback to help shape this release. For detailed guidance on upgrading or using the new features, please refer to the full documentation.

Thank you for using MuSpAn!

On this page