Each model can be run by python <script_name.py>
. The stochastic models can also be run in ensemble through a wrapper.
Found under folder agent_based/
. The simplest possible implementation of an agent-based SIR model, assuming only random mixing between individuals. The population is stored as a Pandas dataframe and can easily be expanded with further characteristics. Very slow to run.
Found under folder compartmental/
.
SIR model based on numerically solving coupled differential equations.
SIR model based on numerically running a set of coupled difference equations with Bernoulli draws of the number of new infections and recoveries at each timestep.