CODEWITHSUNDEEP
pythonpandasdataframesum
Redratz
Redratz

Reputation: 136

Pandas sum of sum of groupby

I'm trying to groupby a column in pandas and then sum the groupby.

Here is an example df and my expected output:

d = {'a':[1, 1, 1, 1, 2, 2, 2], 'b': [3, 4, 5, 6, 7,8,9] }
data = pd.DataFrame(data = d)


# should return sum of sum of groups
# correct output would be 42

I know that I can return the sum of groups by using:

data.groupby('a')['b'].transform(sum)

# which returns

0    18
1    18
2    18
3    18
4    24
5    24
6    24
Name: b, dtype: int64

However, I'm not sure how to get the sum of the sum of groups. i.e.

# sum of groupby
# group 1: 18
# group 2: 24
# sum of sum of groupby
# 18 + 24 = 42

Upvotes: 1

Views: 651

Answers (1)

Pygirl
Pygirl

Reputation: 13349

You need to use agg in place of transform.

res = data.groupby('a')['b'].agg(sum).sum()

res:

42

data.groupby('a')['b'].agg(sum)

will give you

a
1    18
2    24
Name: b, dtype: int64

Upvotes: 2

Related Questions