Network model export

A network generated with Pop2net can be exported as a networkx graph object using model.export_bipartite_network() and model.export_agent_network().

First, let us create a example network:

import networkx as nx

import pop2net as p2n

N_RINGS = 5
N_AGENTS = 25
model = p2n.Model()
creator = p2n.Creator(model=model)
inspector = p2n.NetworkInspector(model=model)


class MyAgent(p2n.Agent):
    def __init__(self, model, *args, **kwargs):
        super().__init__(model, *args, **kwargs)
        self.age = self.model.random.randint(20, 80)


class Ring(p2n.LocationDesigner):
    n_locations = N_RINGS
    nxgraph = nx.cycle_graph(int(N_AGENTS / N_RINGS))

    def weight(self, agent):
        return 10


class Bridge(p2n.LocationDesigner):
    def mutate(self):
        return {"focal_ring_id": list(range(N_RINGS))}

    def filter(self, agent):
        return agent.Ring_id in [self.focal_ring_id, self.focal_ring_id + 1]

    def split(self, agent):
        return agent.Ring_position


creator.create_agents(n=N_AGENTS, agent_class=MyAgent)
creator.create_locations(location_designers=[Ring, Bridge])

LocationList (50 objects)

Let’s have look on the network before we export it:

inspector.plot_networks()

In order to export the bipartite network as a networkx graph object, use model.export_bipartite_network(). Using the arguments location_attrs and agent_attrs, location attributes and agent attributes will be saved as node attributes.

bipartite_graph = model.export_bipartite_network(agent_attrs=["age"])
bipartite_graph

<networkx.classes.graph.Graph at 0x16805efc550>

To transform the network into a agent-level network and save it as a networkx graph object, use model.export_agent_network().

agent_graph = model.export_agent_network(node_attrs=["age"])
agent_graph

<networkx.classes.graph.Graph at 0x1685fdca210>