Continuous space agents and model in 2D

Example: Following code creates 100 continuous space 2D agents, and then a 2D model containing them.

agents = con_2d_agents(100, pos = Vect(5.0,5.0), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_2d_model(agents, force = 2.0, dt=0.1)

Continuous space agents and model in 3D

Example: Following code creates 100 continuous space 3D agents, and then a 3D model containing them.

agents = con_3d_agents(100, pos = Vect(5.0,5.0,5.0), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_3d_model(agents, force = 2.0, dt=0.1)

Discrete space agents and model in 2D

Example: Following code creates 100 discrete space 2D agents, and then a 2D model containing them.

agents = grid_2d_agents(100, pos = Vect(5,5), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_2d_model(agents, size=(20,20) force = 2.0, dt=0.1) # there will be a 20x20 grid of patches. By default a 2d model (discrete as well as continuous) has a 10x10 grid of patches.

Discrete space agents and model in 3D

Example: Following code creates 100 discrete space 3D agents, and then a 3D model containing them.

agents = grid_3d_agents(100, pos = Vect(5,5,5), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_3d_model(agents, size=(20,20,20), force = 2.0, dt=0.1) # there will be a 20x20x20 grid of patches. By default a 3d model (discrete as well as continuous) has a 10x10x10 grid of patches.

Graph space agents and model

Example: Following code creates a 10x10 square grid graph, 100 graph space agents, and a graph space model containing them.

graph = square_grid_graph(10,10)
agents = graph_agents(100, node=1, color = cl"yellow", keeps_record_of = Set([:node])) # agents live on graph nodes, so instead of :pos they have :node property
model = create_graph_model(agents, graph, force = 2.0, dt=0.1) 

Agents type - Mortal vs Static

A model can have Static agents or Mortal agents. Static agents do not reproduce nor die. Mortal agents can die and reproduce. The mortality type of agents can be specified via agents_type argument in the function for creating model and its default value is Static. E.g. in the following code agents_type is set to Mortal.

agents = con_2d_agents(100, pos = Vect(5.0,5.0), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_2d_model(agents, agents_type=Mortal, force = 2.0, dt=0.1) 

Space type - Periodic vs NPeriodic

A 2D or 3D space model can have periodic or non-periodic boundary conditions. This can be specified via space_type argument in the function for creating model and its default value is Periodic. E.g. in the following code space_type is set to NPeriodic.

agents = con_3d_agents(100, pos = Vect(5.0,5.0,5.0), color = cl"yellow", keeps_record_of = Set([:pos]))
model = create_3d_model(agents, space_type=NPeriodic, force = 2.0, dt=0.1) 

Graph space - Dynamic vs Static

A graph can be Dynamic (nodes/edges can be added or removed) or Static (nodes/edges can not be added or removed). In the following code we create a 10x10 dynamic grid graph, create 100 graph agents and then a graph model containing them.

graph = square_grid_graph(10,10, dynamic=true)
agents = graph_agents(100, node=1, color = cl"yellow", keeps_record_of = Set([:node])) # agents live on graph nodes, so instead of :pos they have :node property
model = create_graph_model(agents, graph, force = 2.0, dt=0.1) 

Graph space - Directed vs Simple

In EasyABM we call a graph Directed if its edges are directed, otherwise we call it Simple. In the following code we create a model with a directed graph that has two nodes 1,2 and edge 1->2 between them.

graph = graph_from_dict(
    Dict(
    "num_nodes"=>2,
    "is_directed"=>true,
    "edges"=>[(1,2)]
    )
)
agents = graph_agents(100, node=1, color = cl"yellow", keeps_record_of = Set([:node])) # agents live on graph nodes, so instead of :pos they have :node property
model = create_graph_model(agents, graph, force = 2.0, dt=0.1) 

Initialization, running, visualization, fetching data

After defining agents and the model, the processes of initialization, running, visualization and fetching data are same for all model types. These steps use common functions like init_model!, run_model!, animate_sim, draw_frame, get_agent_data etc. For more details please refer to the api.