Open kyleam opened 2 years ago
@kylebaron I'll defer to you, and any other scientists that you want to weigh in. Do we want to add something like "ofv_nan"
(maybe more descriptive name?) to the heuristics that we return in the summary object, or is it enough just to have OFV come through as NA
(which you would see in the bbr
print method).
Another option, that would be more lightweight, is to only modify the bbr
print method so that it shows some kind of flag in place of the OFV. This would be no changes to the bbi
code.
This builds on top of https://github.com/metrumresearchgroup/bbi/issues/266. Once that lands, there can be
NA_real_
OFV values. This issue is a placeholder to discuss if/how we should alert the user to this bad value (e.g., tacking something on to the heuristics reported in the summary).cc @seth127 @kylebaron