Enables correct results if dataframe contains multiple results for one (application, platform) pair.
Selects the best efficiencies for a given (application, platform) pair to ensure correct results.
I considered making pp and cascade throw an error if the dataframe contained duplicate results, but handling this within the library itself seemed more user-friendly.
Related issues
N/A
Proposed changes
I considered making
pp
andcascade
throw an error if the dataframe contained duplicate results, but handling this within the library itself seemed more user-friendly.