v0.4.0

What's Changed

• Remove clutter from the dsm.Energy class in #17
• Remove dsm.Energy.w attribute in #19
• Bump minimum Python from 3.8 to 3.11 by in #20
• Migrate to repype, task files are now YAML instead of JSON, in #21
• Update to v0.4.0 in #26

Full Changelog: v0.3.0...v0.4.0

v0.3.0

What's Changed

• Add SUPERDSM_NUM_CPUS environment variable by @kostrykin in #6
  • Only for batch processing, defaults to 2
  • The num_cpus setting in task files is no longer supported
• Section about environment variables have been added to the docs: https://superdsm.readthedocs.io/en/latest/usage.html#environment-variables
• Update superdsm.io module by @kostrykin in #10
  • Rename io.imwrite to io.imsave for consistency with other modules
  • Add force_filetype argument for io.imread
• Bump cvxopt from 1.2.6 to 1.2.7 by @dependabot in #3

Full Changelog: v0.2.0...v0.3.0

ISBI 2024 Release (v0.2.0)

What's Changed

• Add the implementation of the pruning method described in Kostrykin and Rohr (ISBI 2024).
• Add the option global-energy-minimization/pruning (see the docs).
• Add examples for reproduction of the results reported in the paper.
• Remove the option global-energy-minimization/strict

Full Changelog: v0.1.3...v0.2.0

Literature

L. Kostrykin and K. Rohr, "Robust Graph Pruning for Efficient Segmentation and Cluster Splitting of Cell Nuclei using Deformable Shape Models," accepted for presentation at IEEE International Symposium on Biomedical Imaging (ISBI), Athens, Greece, May 27–30, 2024.