Output consistency checks required!

[timriffe]

Output consistency checks required!

[timriffe]

a teeny tiny bug. rescale_sexes() wasn't being triggered properly by do_we_rescale_sexes(). Now fixed, re-running processes. You're right, this is a consistency check to run after a processing run.

This constraint happens before PCLM, but I guess it could be lost if age groups are split. Presently, after PCLM there is no re-constraining of this kind. Should we add it?

[mpascariu]

Excellent! It looks much better.
In the example from above there still are some minor differences especially in the 90 and 100 age classes.

   Country Region Code         Date         Age AgeInt       b      f      m   `m+f`
   <fct>   <fct>  <fct>        <fct>      <int>  <int>   <dbl>  <dbl>  <dbl>   <dbl>
10 Spain   All    ES02.05.2020 02.05.2020    90     10  4266.  2486.  1807.   4293. 
11 Spain   All    ES02.05.2020 02.05.2020   100      5   852.   594.   231.    825. 

So far we have written tests for the input data. I think it will not hurt if we can run them over the final output as well.

I can take care of this in a separate script.

[timriffe]

Great, I'll close this issue. FYI, f and m are forced to sum to b now after splitting. Previously they were not. So, moving forward I don't expect this flag to be thrown.