Resettable State

[mmcleod89]

We need to be able to reset the state to the year of intervention. Saving the dataframe should be easy enough, but we need to make sure that nothing in the modules is going to cause a problem e.g. variables that change over time.

[mmcleod89]

How easy is it to serialise custom classes in Python?

[mmcleod89]

We need to be able to retain the state at the time of intervention and then continue with either option 0 (continuation of whatever the status quo is) or option X (some choice of intervention)

[mmcleod89]

In simulation.py

• run simulation to year of intervention instead of end
• make population deep copy and test that these are fully independent i.e. no references, changes in one (either to people or module variables) does not affect the other
• write a simple intervention function and call this from inside run once the copy has been made (and test!)
• run each simulation to stop date and capture output from both