School and workplace layers

[pkasibhatla]

Hi, I am trying to create a population using the "hybrid network" option using the following code:

import covasim as cv
pars = dict(n_agents=1000, pop_type='hybrid')
sim1 = cv.Sim(pars).init_people()

But when I analyze the resulting network, I find that all individuals who are assigned to schools are assigned to the same school and all individuals who are assigned to work places are assigned to the same workplace. Am I doing something wrong?

Thanks for your help.

[cliffckerr]

Hi @pkasibhatla -- that's correct. In the hybrid network, for speed, only households have fully microstructured contacts; school and workplace contacts are age-limited but are not microstructured.

For fully microstructured contacts, you can use pop_type='synthpops'. This requires the SynthPops library, which is no longer actively maintained, but should still work with Covasim.

[pkasibhatla]

Thanks!

[pkasibhatla]

Hi @cliffckerr, I have one additional question - the assumption for both the "hybrid" and the "synthpops" simulations is that the number of community contacts that each individual has is a random draw from a Poisson distribution with mean 20. When this is implemented in a multiday simulation, I assume that community contacts (as well as houshold, workplace, and school contacts) occur on each day of the simulation. In terms of an index case, is the implicit assumption that the index case's community contacts stay the same (in terms of specific individuals) from day to day - or is there a new draw of individuals that comprise the index case's community contacts each day?

Thank you for your help.