net2 0.0.15

Extends 'networkx' for working with spatial networks

net2

Overview

net2 is a Python package for working with spatial networks.

Data structures

net2 is intended to work with spatial network data represented using networkx.DiGraph objects, following several conventions:

• An object of type nx.Graph or nx.DiGraph
• The network is associated with:
  • 'crs' (None or int)
• Node IDs are int
• Nodes are associated with:
  • 'geometry' (shapely)
• Edges are associated with:
  • 'geometry' (shapely)
  • 'length' (float, in $m$)
  • 'time' (float, in $sec$)

Installation

The package can be installed with:

pip install net2

Use

Once installed, the package can be loaded with:

import net2

Sample data

The package comes with a small spatial network for demonstration, accessible as follows:

G = net2.data['roads']
G

<networkx.classes.digraph.DiGraph at 0x7f2bfcedffe0>

Here is a plot of the network:

import matplotlib.pyplot as plt
import networkx as nx
fig, ax = plt.subplots()
nx.draw(G, with_labels=True, pos=net2.pos(G))
plt.axis('on')
ax.set_aspect('equal')
ax.tick_params(left=True, bottom=True, labelleft=True, labelbottom=True);

Example

Calculating the shortest route between two custom points, using net2.route2:

import shapely
pnt1 = shapely.Point(4100, 3100)
pnt2 = shapely.Point(1100, 3700)
x = net2.route2(G, pnt1, pnt2, 'length')
x

Plot of the resulting route and modified network:

fig, ax = plt.subplots()
nx.draw(x['network'], with_labels=True, pos=net2.pos(x['network']), edge_color='grey', width=0.1, ax=ax)
nx.draw_networkx_edges(G, pos=net2.pos(x['network']), edgelist=list(zip(x['route'], x['route'][1:])), edge_color='r', width=3)
plt.plot(pnt1.x, pnt1.y, 'ro')
plt.plot(pnt2.x, pnt2.y, 'ro')
ax.annotate('pnt1', [pnt1.x, pnt1.y], ha='left')
ax.annotate('pnt2', [pnt2.x, pnt2.y], ha='left')
plt.axis('on')
ax.set_aspect('equal')
ax.tick_params(left=True, bottom=True, labelleft=True, labelbottom=True);