1. Data Generation

Data Generation

Our Agent-Based Simulator - Covasim

"An agent-based model is a computational model for simulating the actions and interactions of autonomous agents in order to understand the behavior of a system and what governs its outcomes." - Wikipedia

• How Covasim Works: Covasim is an agent-based simulator that utilizes efficient computations, pseudorandom coding techniques, probability distributions, and in-depth nodal representations of agents and their contact networks to simulate the spread of an infectious disease - COVID-19.

• Our Data Generator: We utilize the files drums_data_gen.py and drums_data_gen_multi.py to generate data.

Within these Python scripts are the interventions (testing and tracing) and behaviors (masking), all of which are instantiations of the type covasim.Interventions class. covasim.Interventions acts as a blueprint class for users to adapt the Covasim in their own way.

• drums_data_gen.py: An old data generator initially used for creating single simulations.

• drums_data_gen_multi.py: The currently used version used for creating any number of simulations. It interacts specifically with the covasim.MultiSim class of type base.FlexPretty, a class that supports multiple different display options. The Sim class is an instance of type base.BaseSim which acts as Covasim's agent-based simulator wrapper class. The Sim class handles the running of the simulation, i.e., the creation of the population and the dynamics of the epidemic and takes in a dictionary of parameters for the simulation. The MultiSim class is for running multiple copies of a Sim instance simulation and takes in the simulation instance that you wish to copy and run multiple times.

Adding an Adaptive Behavior