v0.0.2#

We are pleased to announce the initial public release of SpaceNet, an open-source Python package for the spatial analysis of spatial networks.

SpaceNet provides a collection of methods for analysing the structure, geometry, and spatial characteristics of embedded networks, with a particular focus on supporting methodological development in spatial network analysis. The package offers a consistent API built on top of common scientific Python libraries and is designed for both research and applied workflows.

Global Network Metrics#

  • Computation of global summary statistics for spatial networks, including measures such as network length, alpha index, and mean detour index.

Node Metrics#

  • Functions for computing node-level geometric and topological properties.

Point Pattern Analysis#

The first release introduces the network pair correlation function (netPCF) framework for analysing point patterns defined on spatial networks, including:

  • Univariate netPCF

  • Cross-netPCF

  • Weighted-netPCF for continuous node attributes

Partition methods#

The initial release comes with novel algorithms for node partitions of spatial networks (compact-volume-balanced partitions).

Datasets#

  • Example datasets for tutorials, demonstrations, and reproducible examples.

Utilities#

  • Helper functions for constructing, validating, and manipulating spatial network objects.

Documentation#

The documentation includes:

  • Installation instructions

  • API reference

  • Worked tutorials

  • Example workflows demonstrating the use of the netPCF framework

Acknowledgements#

SpaceNet has been developed to support research into spatial network analysis and the development of new statistical methodology for spatially embedded networks.

We welcome bug reports, feature requests, and community contributions through the project’s GitHub repository.

On this page