You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The make_per_unit function is specific to the replicate function that is being used. In the test cases, the replicate function does deep copy of the single network data and hence, it works fine. If one decides to be memory efficient with the replicate function (by use of pointers), then the make_per_unit implementation would not give the desired effect; for instance, it would normalize pg values by pg/mvaBase^n. I do not know how to address this - but just bringing out the issue.
The text was updated successfully, but these errors were encountered:
If make_per_unit works in the deep-copy case, then I don't think this is a bug. If one is going to use clever space saving methods, then that user needs to be careful with their data manipulation.
A feature we could add to help streamline this would be to have an optional per_unit field in each component. make_per_unit and make_mixed_units could then check this optional value before performing the requested conversion.
The make_per_unit function is specific to the replicate function that is being used. In the test cases, the replicate function does deep copy of the single network data and hence, it works fine. If one decides to be memory efficient with the replicate function (by use of pointers), then the make_per_unit implementation would not give the desired effect; for instance, it would normalize pg values by pg/mvaBase^n. I do not know how to address this - but just bringing out the issue.
The text was updated successfully, but these errors were encountered: