Create `opt_check_tbl_counts_lt_popn()` fn

[annakrystalli]

Predictions are not greater than population size

For flusight, can look up population values from here: https://github.com/cdcepi/FluSight-forecast-hub/blob/main/auxiliary-data/locations.csv

Arguments to this validation function:

• Where to find popn file
• What column in this file has population values
• What column in this file has location values
• Which targets need to be checked? (Need to somehow specify which values (i.e., which rows) need to be checked (combination of task ids and output types?)

Logic:

• Subset to rows that need to be checked
• Join locations.csv and submission file on the specified column (in this example, location)
• If any NAs in popn values, throw error (i. e. Check throws error if submission file contains locations which do not match locations in population file.)
• Check that value less than population in all rows

[annakrystalli]

@elray1 @LucieContamin @kathsherratt Do you use this check and if so, how do you all feel about locations.csv as the default name for the file?

[LucieContamin]

We are using this check on the US SMH, I am ok with the default name of the file. We should just prepare some documentation to go with the file.

[elray1]

agreed, sounds good :)