Currently the nil values in DataFrame are UndefinedObject instance. We would like to replace them by something like DataNaN that could do computations.
For example:
NaN - 5 => NaN
This would remove the need to nil check everywhere and simplify the code
Currently the nil values in DataFrame are UndefinedObject instance. We would like to replace them by something like DataNaN that could do computations.
For example:
NaN - 5 => NaN
This would remove the need to nil check everywhere and simplify the code