Adds a node size plot parameters and cleans up docs

• Adds primal_node_size and dual_node_size parameters to the plot_nX_primal_or_dual function.
• Adds a node_size parameter to the plot_nX method.
• Updates docs.

Flips signs of beta formulations

Flips signs of beta formulations for negative exponential weights with the implication that beta values should now be provided as positive floats instead of negative floats.

Adds OSMnx conversion function.

• Adds a function for converting OSMnx multiDiGraph to cityseer compatible MultiGraph.
• Adds docs providing context on comparison to OSMnx and other graph packages.

Version bump to release status

Bumps version to v1.0.0 so that pre v1 versions are not installed by default via Pip.

v1 alpha release

v1 alpha release

This release includes significant changes including several breaking changes.

Given that the repo is still in alpha / beta development stages these changes won't be discussed here in depth, except to note that:

• The graph cleaning methods have been substantially overhauled. The consolidation methods have changed substantially; see the documentation for examples of how to use the new methods.
• The documentation system has been upgraded: the latest version includes a python docstring parser that builds the markdown from the source code. The static site generator has been migrated from vuepress to vitepress.

The latest version is named for v1 (alpha) indicating that the intent is now to stabilise development going forward.

v0.12.0

Breaking change:

The compute_centrality method has been split into two: compute_node_centrality and compute_segment_centrality. This release incorporates some abstractions pertaining to numba JIT compiled functions and improves angular and segmentised methods.

Version Bumps

v0.11.1

version bumps

v0.11.0

Reverts node betweenness to old behaviour (does not skip ghosted nodes for decomposed graphs).

v0.10.2

• Fixes network map checks.
• Tames temporary numba progress bar.

v0.10.1

Adds a step to shortest-path algorithms to check for self-loop scenarios. (Edges still considered for segment centrality methods.)