HomCloud is a TDA package with both a Python and a CLI interface.
The package can compute persistent homlogy of a wide variety of data, including
- 3D pointcloud analysis
- Binary image analysis
- Grayscale image analysis
- 3D binary image analysis
- Distance matrix analysis (Vietoris-Rips)
- 3D periodic pointcloud analysis
The package also provides a range of histogram-based vectorisation methods, including persistence images and diagram binning. These methods are tightly coupled with the Python machine-learning ecosystem, including scikit-learn.
Utilities for a variety of inverse methods are provided, including
- optimal volumes
- stable volumes
- optimal 1-cycles