Review initial code commit

[JSchoenbachler]

Elliot and I to review Layla's first commit individually.

[JSchoenbachler]

Fixed some minor spacing and formatting issues, beginning code review.

[JSchoenbachler]

• Need function that does everything, only needing to supply the filenames and/or additional parameters for further calculations.
• Some more details on parameters and what is being passed.
• Example data and vignettes.
• .packages in foreach might not be necessary, never used before.
• Add imports for foreach functions and fixes for the data.table column name issues with running R CMD check.
• Probably need to add some assertions to make sure data being used is valid.
• Update DESCRIPTION with imports and further detail.
• Might want to rename the functions to better fit in as being the "original" PheRS model.
• Add a README for clarity.
• Getting weightsAll and the subsequent merge seem like there might be some unnecessary (or rather inefficient) processing being done. For instance, if we are setting dx_status as 0 if is.na(phecode) and the subsequent weight calculation is therefore 0, why are we using all.x = TRUE in the merge for weightsSub? I think this could be revisited.
• For that matter, making weightsSub could probably be a function of it's own. I know it isn't re-used anywhere currently, but it might just be useful to have it broken up. I'm undecided on this though.

[ehoutland]

• In calcWeights, I think you need to specify data.table::uniqueN since you're not importing the whole package.
• Do you guys think the original and age-aware PheRS should be different functions entirely or just be different options in the same function? I know that's not what we're doing right now, just wondering.
• The parameter values should say what's being passed and describe default behavior.
• Small thing, but you can just use single quotes for everything instead of alternating with double quotes.
• Since calcPheRS takes GRIDs and phecodes, it seems like it should just call calcWeights on its own instead of the weights having to be provided separately. Maybe if the weights argument is missing it could do that?

[JSchoenbachler]

I'm marking this first code review as "Done" on the project, but keeping the Issue set as "Open" for ease of access while work is being done on the package.