if I want to join the duplicated some columns in dataframe?

Question

col1 col2 col3
jake  male   180
jake  male    140
john   female 230
john   male    100

let say we have this kind of df
if I want to group col1, and col2 and make new dataframe like this

col1 col2 col3
jake  male  [180,140]<-mean
john  female [230]<-mean
john  male   [100]<-mean

I tryied to use the group by

keyword_mean = df.groupby(['col1','col2'])['col3'].agg(['mean']).reset_index()

but it keep making the duplicated col1 and col2

Answer 1

If you're looking for a pd.Series of the means without any group info, try:

df.groupby(["col1", "col2"]).mean().reset_index()["col3"]