Output agent-level information

[ingmarschuster]

Hi,

would it be possible to output agent-level adjacency matrices (capturing contacts between agents) and SEIR status per time point?
This would be super useful to check advanced testing schemes that have appeared in the literature.

Thanks a lot!
Ingmar

[syshafi]

Hi Ingmar,
Thank you for your issue, and I'm sorry about the slow reply. It should be possible to output summary states by configuration already, and we will shortly finish functionality to output a notion of agent-level adjacency matrices for the risk_learning application. I will add an example in the next several days, but the parameters of interest are here:
https://github.com/google-research/agent-based-epidemic-sim/blob/develop/agent_based_epidemic_sim/applications/risk_learning/config.proto#L181 and
https://github.com/google-research/agent-based-epidemic-sim/blob/develop/agent_based_epidemic_sim/applications/risk_learning/config.proto#L182

Note that you will need to instantiate files containing AgentProto and LocationProto messages in order to run the risk learning simulation.

You can also use the contact_tracing application to do this. Take a look at https://github.com/google-research/agent-based-epidemic-sim/blob/develop/agent_based_epidemic_sim/configuration_generator/config.py to generate a feasible configuration.