Package modelarium.agents

package modelarium.agents

Agent abstractions used in Modelarium simulations.

This package defines the core concepts for modelling autonomous agents within a simulation. Agents encapsulate state and behaviour, and are typically updated on each simulation step by the engine.

Implementations in this package are intended to be extended by users to define domain-specific agent logic, such as decision-making, interactions with the environment, or state transitions over time.

The agent lifecycle and update frequency are controlled by the surrounding simulation and scheduling components, rather than by agents themselves.

Related Packages

Package	Description
modelarium	Modelarium API.
modelarium.attributes	Attribute abstractions for storing and accessing model state.
modelarium.environments	Environment abstractions for Modelarium simulations.
modelarium.multithreading	Internal multithreading and parallel execution support.
modelarium.results	Result recording and retrieval for Modelarium simulations.
modelarium.scheduler	Scheduling and execution control for Modelarium simulations.

Classes

Class	Description
Agent	Represents an agent in the agent-based model.
AgentGenerator	Abstract base class for generating agent populations in a simulation.
AgentSet	A collection class for managing Agent instances, with support for: Optional deep copying of agents on insertion Fast lookup by agent name Filtering, duplication, and setup routines Randomised iteration
DefaultAgentGenerator	A basic implementation of AgentGenerator that creates agents with deep-copied base attribute sets and assigns them unique sequential names.
FunctionalAgentGenerator	An implementation of AgentGenerator that delegates agent creation logic to a user-defined functional interface.