Its possible to slice my data where I have the main key for all child codes, considering the main code will be duplicate in some moments, so i have year column to support the data slice.

Thats my current data and I need to transform it

And that is what i looking

CodePudding user response：

I'm not clear what you are asking here. Your example appears to be longer than after slicing. are you perhaps looking to re-order the data?

a multi-index perhaps?

CodePudding user response：

If I have this correct I think you are looking to get a row in the data frame with every possible combiantion of the data.

Also eliminate spaces in column headers is a good habit and will save a lot of time. The code looks cleaners and faster to type

df = df.rename(columns={'child code' : 'child_code'})

df2=pd.DataFrame(columns=['cod','child_code', 'year','text'])


for item1 in df.cod.unique():
    for item2 in df.child_code.unique():
        for item3 in df.year.unique():
            for item4 in df.text.unique():
                new_row = pd.DataFrame({'cod' : [item1], 'child_code' : [item2], 'year' : [item3],'text': [item4]})
                df2 = pd.concat([df, df2])```