Module specification

(still writing, more too come..)

R interface basics

The python code will call you're R scripts, setting some variable in the environment:

• from_python: truthy boolean, test for this to know if your code is run automatically.
• ti_str, tf_str model start and end as a string
• foldername the folder that contains everything related to the setup. You'll have to load geodata.csv from there. It include the / at the end.

if (!from_python) {         # or whatever values you use to test.
    ti_str <- '2020-01-31'
    tf_str <- '2020-08-31'
    foldername <- 'west-coast/'      
}
# write code here that uses what is above and can load more files.

The code is run from the root folder of the repository.

Setup

A setup has a name, and this name is a also folder that contains file geodata.csv (see below).

Modules

(and status if the current R implementation respect the specification)

Mobility (WIP)

• From R: dataframe named mobility with columns: from, to, amount. Relationships not specified will be set to zero. You can set different value for A -> B and B -> A (if you only specified A -> B, we'll assume B -> A = 0).
• From file: matrix to be imported with numpy as it is. Dimension: (nnodes, nnnodes) (may have a third dimension if time varying). First index is from, second is to, diagonal is zero (mobility[ori, dest])
• From python: numpy matrix as file.

Population (DONE)

• From file: geodata.csv : specification of the spatial nodes, with at least column for the zero based index, the geoid or name, the population.

Importation (TODO)

• From R: dataframe named importation with column date, to, amount where date is a string, to contains a geoid and amount contains an integer.

NPI (DONE)

Transmission parameters (TODO)