Pandas groupby two columns and get max value

Question

Pandas groupby two columns and get max value

I have grouped data with multiindex

                        
    Model  VehicleType  VehicleType            
    100    sedan                278
           wagon                109
           coupe                  2
           convertible            1
    145    small                 19
    ...                         ...
    zafira sedan                 22
           small                 11
           suv                    7
           convertible            1
           coupe                  1

I need to get max value of count (right column) with corresponding model and VehicleType, like this:

                        
    Model  VehicleType  VehicleType            
    100    sedan                278
    145    small                 19
    ...                         ...
    zafira sedan                 22
           

Thanks for solutions!

Answer 1

Use DataFrameGroupBy.idxmax for indices by maximum value and then select by DataFrame.loc :

df = df.loc[df.groupby(level=0)['VehicleType'].idxmax()]
print (df)
                    VehicleType
Model  VehicleType             
100    sedan                278
145    small                 19
zafira sedan                 22