weekly aggregation of cases

[sbfnk]

This is done by default at the moment and the user can't look at the underlying simulated data - wouldn't it make more sense for outbreak_model() to return the full simulation and have a separate function for aggregating to weekly?

[pearsonca]

agree - not like there's a performance reason, since it's all computed at daily already and the memory allocation already gets paid.

[joshwlambert]

I agree hard coding the weekly aggregation is not ideal. But instead of splitting and removing the aggregation into its own exported function, my preference would be to add an aggregation interval to sim_opts() to allow, daily, weekly, etc, or none, similar to {incidence2}.

If we wanted or users requested the aggregation available after the simulation in a future version it would be as simple as moving the aggregation function from internal to exported.