Phimaker (Persistent homology of images and (co)kernels) computes the persistent homology of kernels, images and cokernels for an inclusion of filtrations, using the algorithm of Cohen-Steiner, Edelsbrunner, Harer and Morozov. It is written in Rust, with Python bindings provided via PyO3.
Persistence computations are lock-free and parallel, built on the LoPHAT library. Phimaker is also used as the persistence backend of Chalc for computing 6-packs of chromatic point clouds.
Paper the implementation is based on:
- Cohen-Steiner, D., Edelsbrunner, H., Harer, J. and Morozov, D. Persistent homology for kernels, images, and cokernels. Proceedings of the Twentieth Annual ACM-SIAM Symposium on Discrete Algorithms (SODA 2009), pp. 1011–1020.
The package can be installed with
pip install phimaker