Predefined Models

Predefined agent based models exist in the Models submodule in the form of functions that return model, agent_step!, model_step! when called.

They are accessed like:

using Agents
model, agent_step!, model_step! = Models.flocking(; kwargs...)

The Examples section of the docs outline how to use and interact with each model.

Warn

Please notice that the predefined models are a convenience and not considered part of the public API. This means that they can have breaking changes between versions of Agents.jl without warning.

So far, the predefined models that exist in the Models sub-module are:

Agents.Models.flocking — Method

flocking(;
    n_birds = 100,
    speed = 1.0,
    cohere_factor = 0.25,
    separation = 4.0,
    separate_factor = 0.25,
    match_factor = 0.01,
    visual_distance = 5.0,
    extent = (100, 100),
    spacing = visual_distance / 1.5
)

Same as in Flocking model.

source

Agents.Models.schelling — Method

schelling(;
    numagents = 320,
    griddims = (20, 20),
    min_to_be_happy = 3,
)

Same as in Schelling's segregation model.

source

Agents.Models.sir — Method

sir(;
    C = 8,
    max_travel_rate = 0.01,
    Ns = rand(50:5000, C),
    β_und = rand(0.3:0.02:0.6, C),
    β_det = β_und ./ 10,
    infection_period = 30,
    reinfection_probability = 0.05,
    detection_time = 14,
    death_rate = 0.02,
    Is = [zeros(Int, length(Ns) - 1)..., 1],
    seed = 19,
)

Same as in SIR model for the spread of COVID-19.

source

Agents.Models.zombies — Method

zombies(; seed = 1234)

Same as in the Zombie Outbreak example.

source