groupby cast to int where possible

[mikisDeWitte]

Hello,

I need to create a nested dictionary like so:
d= {"ftsg2a_co422": {1: 'visited or contacted', 0: 'not visited not contacted'}}

this dictionary then goes to a function that sets value labels to an SPSS file.

The data comes from a database where these 0's and 1's can be a string.
When I execute the code below, I get
'1': 'promotor'

Is there a way where I can modify this groupby, cast the value to integer where possible, so that I get
1 : promotor
'rgDEU': 'Germany'
etc?

I hope my question makes sense...

Regards,
Mikis

import pandas as pd
#So all of this is read in from a database:
var= ["ftq10a_bra1", "ftq10a_bra1", "ftq10a_bra1", "ftq10a_bra2", "ftq10a_bra2", "ftq10a_bra2", "d1ccode", "d1ccode"]
value= ["1", "2", "3", "1", "2", "3", "rgDEU", "rgFRA"]
label= ["promotor", "passive", "detractor", "promotor", "passive", "detractor", "Germany", "France"]
data_tuples = list(zip(var,value, label))
column_labels = pd.DataFrame(data_tuples, columns=['var','value', 'label']) 
   
#This is what I need to modify, so that "1" becomes 1
d = {k: f.groupby('value')['label'].apply(lambda x: x.iloc[0]).to_dict()
     for k, f in column_labels.groupby('var')}