How to agg with 2 different parameters-CodePudding

im having a trouble with some work. I need to make these:

tarjeta_maestro.groupby(by=['Cod Marca Producto', 'Cod Comercio']).agg({'Nro Liquidacion': 'max',
                                                                        'VENTAS C/DESCUENTO CONTADO': 'sum',
                                                                        'ARANCEL': 'sum', 
                                                                        'IVA CRED.FISC.COMERCIO S/ARANC 21,00%':'sum', 
                                                                        'Iva 21':'sum',
                                                                        'RETENCION IVA': 'sum',
                                                                        'PER B.A.I.BR.DN.01/04': 'sum', 
                                                                        'RETENCION IMP.GANANCIAS': 'sum',
                                                                        'RETENCION ING.BRUTOS': 'sum',
                                                                        'RETENCION ING.BRUTOS SIRTAC': 'sum',
                                                                        'TOTAL DEDUCCIONES': 'sum',
                                                                        'TOTAL LIQUIDACION': 'sum',
                                                                        'Diferencia' : 'sum' })

My problem is that I need to pass a parameter 'max', and the rest of the columns 'sum', but I can't find a way to achieve it without specifying the function corresponding to each column. I need to apply this to several more df with different columns I imagine that it should be possible to do it in a simple way, but I couldn't find the solution. Could you guide me to solve this problem?

CodePudding user response：

You can use dictionary comprehension to generate the agg function

# You can access desired columns with df.columns
# if you have many target columns and they are continuous
columns = ['Nro Liquidacion', 'VENTAS C/DESCUENTO CONTADO', 'ARANCEL', 'IVA CRED.FISC.COMERCIO S/ARANC 21,00%', 'Iva 21', 'RETENCION IVA', 'PER B.A.I.BR.DN.01/04', 'RETENCION IMP.GANANCIAS', 'RETENCION ING.BRUTOS', 'RETENCION ING.BRUTOS SIRTAC', 'TOTAL DEDUCCIONES', 'TOTAL LIQUIDACION', 'Diferencia']
max_columns = ['Nro Liquidacion']

d = {col: ('max' if col in max_columns else 'sum') for col in columns}

tarjeta_maestro.groupby(by=['Cod Marca Producto', 'Cod Comercio']).agg(d)

CodePudding user response：

You can insert tarjeta_maestro.columns.drop('Nro Liquidacion')[0] : 'sum' inside agg:

tarjeta_maestro.groupby(by=['Cod Marca Producto', 'Cod Comercio']).\
agg({'Nro Liquidacion': 'max',\
    tarjeta_maestro.columns.drop('Nro Liquidacion')[0] : 'sum'})