GUDHI version 3.12.0

We are pleased to announce the release 3.12.0 of the GUDHI library.

As a major new feature, the GUDHI library now offers Čech complex persistence scikit-learn like interface. Python bindings were rewritten using NanoBind.

The GUDHI library is mainly developped using GitHub, do not hesitate to fork the GUDHI project on GitHub. From a user point of view, we recommend to download GUDHI user version (gudhi.3.X.X.tar.gz).

Below is a list of changes:

• Čech complex [Python]
  • Čech complex persistence scikit-learn like interface provides 2 new classes:
    • CechPersistence for computing the same persistent homology as the Čech complex, while being significantly smaller, using internally a DelaunayCechComplex.
    • WeightedCechPersistence for computing the same persistent homology as the Weighted Čech complex, while being significantly smaller, using internally a Weighted version of AlphaComplex.
• Alpha complex dD [C++]
  • API break: The simplicial complex for the Alpha complex concept has been changed to dimension_simplex_range (that must returns a range of simplices of a given dimension) instead of skeleton_simplex_range (that was returning a range of simplices lower or equal to a given dimension)
• Simplex_tree [C++]
  • A new iterator over the simplices of the simplicial complex that match a given dimension
• Persistence matrix [C++]

  Matrix API is in a beta version and may change in incompatible ways in the near future.

  • A new function to insert a maximal cell using swaps.
• Representations [Python]
  • in metrics, BottleneckDistance argument epsilon is deprecrated and renamed e to be consistent with bottleneck_distance and pairwise_persistence_diagram_distances
• Installation
  • Python bindings are now requiring NanoBind. PyBind11 and Cython dependencies are no more required.
  • Minimal Python version is now ≥ 3.10.0 (was ≥ 3.9.0).
  • Minimal NumPy version is now ≥ 1.21.6 (was ≥ 1.15.0).
  • Pip package is now available for OSx ≥ 13.0 (was ≥ 12.0).
  • Support Eigen 3.4.1 and 5.X.X
• Miscellaneous
  • The list of bugs that were solved is available on GitHub.

All modules are distributed under the terms of the MIT license. However, there are still GPL dependencies for many modules. We invite you to check our license dedicated web page for further details.

We kindly ask users to cite the GUDHI library as appropriately as possible in their papers, and to mention the use of the GUDHI library on the web pages of their projects using GUDHI and provide us with links to these web pages.

We provide bibtex entries for the modules of the User and Reference Manual, as well as for publications directly related to the GUDHI library.

Feel free to contact us in case you have any questions or remarks.

For further information about downloading and installing the library (C++ or Python), please visit the GUDHI web site.