Skip to content

Latest commit

 

History

History
34 lines (31 loc) · 935 Bytes

homcloud.md

File metadata and controls

34 lines (31 loc) · 935 Bytes
name title tags links
homcloud
HomCloud
lang/python
type/persistence
type/representative
complex/simplicial
complex/cubical
complex/rips
custom/distance
repr/image
name url
Homepage

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