CODEWITHSUNDEEP
pythonpandasdataframegroup-by
seb
seb

Reputation: 11

Compare every value of dataframe

I have a data with 1000 rows and 20 columns, and I would like to group by value some row.

Here an example of what I have :


column A   column B  column C  column D   column E

XX        YY         25/01/2022   25         50
XX        YY         17/01/2023   32         10

I would like group by per column A and column B and sum value of column D and column E

Here an example of what I want :

column A   column B  column C  column D   column E

XX        YY         25/01/2022   57        60

I kept the inferior date . I would like do it in python with pandas maybe.

Thanks for reading me.

if I need group by only when column A is XX, what could I add for this ?

Upvotes: 1

Views: 43

Answers (1)

T C Molenaar
T C Molenaar

Reputation: 3260

Use df.groupby() and aggregation first for column C and sum for columns D and E.

data = {'A': ['XX', 'XX'], 'B': ['YY', 'YY'],
        'C':['25/01/2022', '17/01/2023'],
        'D': [25, 32], 'E': [50, 10]}
df = pd.DataFrame(data)
df[df.A == 'XX'].groupby(['A', 'B'], as_index=False).agg({'C':'first','D':'sum','E':'sum'})

Output:

    A   B           C   D   E
0  XX  YY  25/01/2022  57  60

Upvotes: 1

Related Questions