0

I am trying to kreate a nested loop to calculate the price combinations between the Car ModelName, e.g.: 

fact_date     CarID   Country  Type   ManufactureNameSum      Sum_Costs
2017-07-14    1       USA      Car    Ford_Focus____VW_Jetta  34003

which means I have a combination of the ManufactorName and ModelName and the sum of the AcquisitionPrice. Somehow I receive the results twice due to the nested loop but just want to have it once. I was thinking solving it with the following line:

row['ManufactureName'] != row2['ManufactureName']

but this only solves the problem with the same ManufactureName but not the one having due to two lists twice the result.

Hope somehone could help here.

Adjustment

I adjusted the code for the nested loop, but still have the propblem concerning the fact_date indexing as I had before.

Adjusted Code:

import pandas as pd

df = pd.read_csv('C:/Sales_Cars.csv', encoding='cp1252', sep=';', index_col=0).dropna()
df2 = pd.DataFrame([])
for current_date in df.index.unique():
    for i in range(0, len(df)):
        for j in range(i+1, len(df)):
            if (
                    df.iloc[i]['ManufactureName'] != df.iloc[j]['ManufactureName'] and
                    df.iloc[i]['CarID'] == df.iloc[j]['CarID'] and
                    df.iloc[i]['Country'] == df.iloc[j]['Country']):
                df2 = df2.append(
                    pd.DataFrame({
                        'CarID': df.iloc[i]['CarID'],
                        'Country': df.iloc[i]['Country'],
                        'Type': df.iloc[i]['Type'],
                        'ManufactureNameSum': (
                            df.iloc[i]['ManufactureName'] +
                            '_' +
                            df.iloc[i]['ModelName'] +
                            '____' +
                            df.iloc[j]['ManufactureName'] +
                            '_'+df.iloc[j]['ModelName']
                        ),
                        'Sum_Costs': (
                            df.iloc[i]['AcquisitionPrice'] +
                            df.iloc[j]['AcquisitionPrice']
                        )
                    }, index=[0]), ignore_index=True) 
print(df2)

The data looks like the following:

fact_date CarID   ManufactureName ModelName    Type   Country AcquisitionPrice
2017-07-14    1   Ford            Focus        Car    USA 11640
2017-07-14    2   Ford            Mustang      Car    USA 12994
2017-07-14    3   Ford            Fiesta       Car    USA 12842
2017-07-14    4   Ford            Mondeo       Car    USA 14685
2017-07-14    1   VW              Jetta        Car    USA 22363
2017-07-14    2   VW              Polo         Car    USA 20107
2017-07-14    3   VW              Golf         Car    USA 21256
2017-07-14    4   VW              Parteon      Car    USA 23679
2017-07-14    1   Toyota          Prius        Car    USA 14384
2017-07-14    2   Toyota          Avensis      Car    USA 14821
2017-07-14    3   Toyota          Corolla      Car    USA 12480
2017-07-14    4   Toyota          Land Cruiser Car    USA 11502
2017-07-14    1   BMW             1er          Car    USA 35127
2017-07-14    2   BMW             2er          Car    USA 43924
2017-07-14    3   BMW             3er          Car    USA 40573
2017-07-14    4   BMW             4er          Car    USA 36690
2017-07-14    1   Mercedes        C-Klasse     Car    USA 36646
2017-07-14    2   Mercedes        A-Klasse     Car    USA 40912
2017-07-14    3   Mercedes        B-Klasse     Car    USA 39060
2017-07-14    4   Mercedes        E-Klasse     Car    USA 41838
Improve this question
12
  • Use ` itertools.combinations` to generate the unique pairs of indices to iterate over Commented Aug 31, 2018 at 9:59
  • @kevinkayaks, do you have an example based on my problem? Is it not possible via nested loop? Commented Aug 31, 2018 at 10:00
  • If you're set on the double loop, you should start the second loop at the value of the index in the first loop I think. Consider this post stackoverflow.com/questions/29324025/… Commented Aug 31, 2018 at 10:11
  • But I would generate the required index pairs with itertools.combinations and use one loop, personally Commented Aug 31, 2018 at 10:13
  • @kevinkayaks, good point starting in the second loop the index like i+1. How could I perfomr it in my example? Commented Aug 31, 2018 at 10:22
Related questions
1
How to avoid nested 'for loops' in Python ?
0
Nested for loops only run once
1
Nested loop in python dataframe
Load 7 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.