CODEWITHSUNDEEP
pythonpandas
jf328
jf328

Reputation: 7351

pandas, groupby.agg to return with a given column name

Currently, I do

df.groupby('A')['B'].agg({'total' : np.sum})

so that the aggregated column has a column name total. But recently I get a warning

FutureWarning: using a dict on a Series for aggregation is deprecated and will be removed in a future version

So what is the preferred way now?

Upvotes: 3

Views: 9679

Answers (2)

jezrael
jezrael

Reputation: 863256

I think you need rename or define name parameter in reset_index:

Check also deprecate groupby agg with a dictionary when renaming.

1.

df = df.groupby('A')['B'].sum().reset_index(name='total')

2.

df = df.groupby('A', as_index=False)['B'].sum().rename(columns={'B':'total'})

3.

df = df.groupby('A').agg({'B' : 'sum'}).rename(columns={'B':'total'}).reset_index()

Upvotes: 3

Zero
Zero

Reputation: 76967

Here are some ways

In [57]: df.groupby('A')['B'].sum().to_frame('total')
Out[57]:
   total
A
a      2
b      1

In [58]: df.groupby('A')['B'].agg(np.sum).to_frame('total')   # or agg('sum')
Out[58]:
   total
A
a      2
b      1

In [59]: df.groupby('A').agg({'B': np.sum}).rename(columns={'B': 'total'})
Out[59]:
   total
A
a      2
b      1

Upvotes: 3

Related Questions