Sum column's values from duplicate rows python3 - sum

Question

I have a old.csv like this:
Name,State,Brand,Model,Price
Adam,MO,Toyota,RV4,26500
Berry,KS,Toyota,Camry,18000
Berry,KS,Toyota,Camry,12000
Kavin,CA,Ford,F150,23000
Yuke,OR,Nissan,Murano,31000
and I need a new.csv like this:
Name,State,Brand,Model,Price
Adam,MO,Toyota,RV4,26500
Berry,KS,Toyota,Camry,30000
Kavin,CA,Ford,F150,23000
Yuke,OR,Nissan,Murano,31000
As you can see the difference from these two is:
Berry,KS,Toyota,Camry,18000
Berry,KS,Toyota,Camry,12000
merge to
Berry,KS,Toyota,Camry,30000
Here is my code:
import pandas as pd
df=pd.read_csv('old.csv')
df1=df.sort_values('Name').groupby('Name','State','Brand','Model')
.agg({'Name':'first','Price':'sum'})
print(df1[['Name','State','Brand','Model','Price']])
and It didn't work,and I got these error:
File "------\venv\lib\site-packages\pandas\core\frame.py", line 4421, in sort_values stacklevel=stacklevel)
File "------- \venv\lib\site-packages\pandas\core\generic.py", line 1382, in _get_label_or_level_values raise KeyError(key)
KeyError: 'Name'
I am a totally new of python,and I found a solutions in stackoverflow:
Sum values from Duplicated rows
The site above has similar question as mine,But It's a sql code,
Not Python
Any help will be great appreciation....

Answer 1

import pandas as pd
df = pd.read_csv('old.csv')
Group by 4 fields('Name', 'State', 'Brand', 'Model') and select Price column and apply aggregate sum to it,
df1 = df.groupby(['Name', 'State', 'Brand', 'Model'])['Price'].agg(['sum'])
print(df1)
This will give you a required output,
sum
Name State Brand Model
Adam MO Toyota RV4 26500
Berry KS Toyota Camry 30000
Kavin CA Ford F150 23000
Yuke OR Nissan Murano 31000
Note: There is only column sum in this df1. All 4 other columns are indexes so to convert it into csv, we first need to convert these 4 index columns to dataframe columns.
list(df1['sum'].index.get_level_values('Name')) will give you an output like this,
['Adam', 'Berry', 'Kavin', 'Yuke']
Now, for all indexes, do this,
df2 = pd.DataFrame()
cols = ['Name', 'State', 'Brand', 'Model']
for col in cols:
df2[col] = list(df1['sum'].index.get_level_values(col))
df2['Price'] = df1['sum'].values
Now, just write df2 to excel file like this,
df2.to_csv('new.csv', index = False)