search

globaldothealth / monkeypox

Mpox 2022 repository
Other
175 stars 36 forks source link

G.h to CDC, WHO comparisons #165

Closed jim-sheldon closed 2 years ago

jim-sheldon commented 2 years ago

Added script to pull data, compare to G.h, update sheet Renamed folders, added directory for templating utility scripts

abhidg commented 2 years ago

@tvarrelman This script currently assumes that US cases are fungible for a particular state, is that likely to continue? I am thinking of a scenario where there might be info added to a particular case (symptoms); and CDC reports two less cases than us, and we set these two to omit_error, one of which is actually confirmed and has more info such as symptoms. OTOH, if all US cases now, only have location, date and status, then this is OK

jim-sheldon commented 2 years ago

Looks OK, I'd add some tests to check that updating the spreadsheet works correctly

Agreed, I am working on testing it now.

jim-sheldon commented 2 years ago

@tvarrelman This script currently assumes that US cases are fungible for a particular state, is that likely to continue? I am thinking of a scenario where there might be info added to a particular case (symptoms); and CDC reports two less cases than us, and we set these two to omit_error, one of which is actually confirmed and has more info such as symptoms. OTOH, if all US cases now, only have location, date and status, then this is OK

This is a great point. Should I add any checks for other columns like symptoms, and only change cases to "omit_error" when those columns are empty?

tvarrelman commented 2 years ago

@tvarrelman This script currently assumes that US cases are fungible for a particular state, is that likely to continue? I am thinking of a scenario where there might be info added to a particular case (symptoms); and CDC reports two less cases than us, and we set these two to omit_error, one of which is actually confirmed and has more info such as symptoms. OTOH, if all US cases now, only have location, date and status, then this is OK

This is a great point. Should I add any checks for other columns like symptoms, and only change cases to "omit_error" when those columns are empty?

Great point @abhidg. Let me take a look at some cases where meta data has been backfilled.

jim-sheldon commented 2 years ago

I refactored scripts into their own directory with shared building and testing strategies and files. It should now be easy to add new ones in a new subdirectory. In the future we could add a tool like cookiecutter to make it even easier.