Model

class Model(starttime=0, stoptime=0, dt=1, name='', scheduler=None, data_collector=None)

This is the main agent base / System dynamics / Hybrid model class It can run manually generated SD models, AB Models or define hybrid models.

add_equation(equation, lambda_method)

Add an equation. ALWAYS use this method to configure equations! Configures the memo as well!

param equation

Name of the equation

param lambda_method

A lambda function we can insert into the set of equations

return

None

agent(agent_id)

Receive one agent by ID

param agent_id

ID of agent (int)

return

Agent object

agent_count(agent_type)

Get number of agents

param agent_type

Agent type to get count for

return

Number of agents (int)

agent_count_per_state(agent_type, state)

Get number of agents in a specific state

param agent_type

Agent type to get count for

param state

state of agents to get count for

return

Integer

agent_ids(agent_type)

Receive agent ids for all agents of agent_type

param agent_type

agent type to get ids for

return

List

begin_episode(episode_no)

When running a simulation repeatedly in episodes, this method is called by the framework to allow tidy up at the beginning of an episode, e.g. a “soft” reset of the simulation. The default implementation calls begin_episode on each agent.

param episode_no

the number of the episode

return

None

begin_round(time, sim_round, step)

Should be called by a scheduler at the beginning of each round, before the agents act methods are called. Add any logic here that is needed to update dynamic properties.

param time

t

param sim_round

round number

param step

step number of round

return

None

biflow(name)

Create a SD biflow

param name

Name of the biflow

return

Biflow object

broadcast_event(agent_type, event_factory)

Broadcast an event to all agents of a particular agent_type

param agent_type

Agent types that are to receive the event

param event_factory

event factory (function) that creates an appropriate event for a given target agent_id

return

configure(config)

Called to configure the model using a dictionary, which itself typically comes from a config file.

param config

Configuration dictionary

return

None

constant(name)

Create a SD constant

param name

Name of the constant

return

Constant object

converter(name)

Create a converter

param name

Name of the converter

return

Converter object

create_agent(agent_type, agent_properties)

Create one agent

param agent_type

Type of agent

return

None

create_agents(agent_spec)

Create agents

param agent_spec

Specification of Agent (dictionary)

return

None

end_episode(episode_no)

When running a simulation repeatedly in episodes, this method is called by the framework to allow tidy up at the end of an episode. The default implementation calls end_episode on each agent.

param episode_no

the number of the episode

return

None

end_round(time, sim_round, step)

Should be called by a scheduler at the end of each round, before the agents act methods are called. Add any logic here that is needed to update dynamic properties.

param time

t

param sim_round

round number

param step

step number of round

return

None

enqueue_event(event)

Called by the framework to enqueue events. In general you don’t need to override this method.

param event

Event instance

return

None

equation(equation, arg)

This method only exists for making SD-DSL models compatible with the new SDModel equation API. In later steps, we might want to extend this method to allow for arrayed equations as well! However, first the DSL needs to support arrays as well!

param equation

equaiton name

param arg

t

property equation_prefix

An id that is unique within this model that can be used to generate unique equation names :return:

evaluate_equation(name, t)

Evaluate an element’s equation

param name

Name of the equation

param t

timestep to evaluate for

return

float of simulation result

flow(name)

Create a SD flow

param name

Name of the flow

return

Flow object

function(name, fn)

Create a user defined function for SD models. :param name: name of the function :param fn: returns :return: a nary function that creates a NaryFunction class

get_property(name)

Get one property

param name

Name of property

return

Dictionary for property

static get_random_integer(min_value, max_value)

Just compute a random integer within bounds

param min_value

min value for random integer

param max_value

max value for random integer

return

Integer

instantiate_model()

Instantiate model stub. Called directly after the model is instantiated. Implement this method in your model to perform any kind of initialization you may need.

return

None

lookup(x, points)

Lookup function: Interpolate between set of points. E.g. for “graphical functions” as known from SD

param x

x-value to find the y value for

param points

return

memoize(equation, arg)

Memoize method - used by the system dynamics equations to remember values that have already been calculated.

param equation

name of equation

param arg

argument (t)

return

result of equation

next_agent(agent_type, state)

Get the next agent by type and state.

param agent_type

Agent type

param state

State the agent is in

return

Agent object

random_agents(agent_type, num_agents)

Receive a number of random agents

param agent_type

Type of agent

param num_agents

Number of agents to receive

return

Agent IDs list

random_events(agent_type, num_agents, event_factory)

Distribute a number of random events

param agent_type

Agent types supposed to receive events

param num_agents

Number of random agents

param event_factory

event factory (function) that creates an appropriate event for a given target agent_id

return

None.

register_agent_factory(agent_type, agent_factory)

Register an agent factory

param agent_type

Type of agent to register

param agent_factory

Function (typically lambda, but not limited to). Input: agent_id, model -> Output: Agent of agent_type

return

None

reset()

Reset simulation

return

None

reset_cache()

Reset memo of all equations

return

None

run(show_progress_widget=False, collect_data=True)

Initiate simulation - this esssentially just calls the run method of the models scheduler.

param show_progress_widget

Boolean: If true, shows a progress widget (only in Jupyter environment!)

return

None

run_specs(starttime, stoptime, dt)

Configure the runspecs of the model.

param starttime

The starttime of the model.

param stoptime

The stoptime of the model.

param dt

The dt of the model.

return

None

set_property(name, property_spec)

Configure a property of the simulation

param property_spec

Specification of property (dictionary)

return

set_scenario_manager(scenario_manager)

Set the scenario manager name

param scenario_manager

String

return

None

statistics()

Get statistics from DataCollector

return

None

stock(name)

Create a SD stock

param name

name of the stock

return

Stock object