Combine dataframe values based on repeated index labels

Question

I've got a single-column dataframe with an index of integers represented as strings that has repeated values in it. The values are integers and I would like to have a dataframe with an index with no repeats in it and whose values are the sum of all the values that originally had the given index label. Here's a sample of the data I'm working with:

>>>  verts
3    54
3    34
0    33
4    28
4    23
2    22
2    15
5    15
5    15
0     9
1     2
6     1
1     1
6     1

I could do it this way, but it doesn't seem like good pandas syntax:

new_index = set(verts.index)
new_vals = [verts[x].sum() for x in new_index]
new_df = pd.DataFrame({'Counts': new_vals}, index=new_index)
new_df
   Counts
1       3
0      42
3      88
2      37
5      30
4      51
6       2

Is there something more straight-forward? Thanks.

Answer 1

Try resetting your index and then using groupby :

verts = pd.Series([54, 34, 33, 28, 23, 22, 15, 15, 15, 9, 2, 1, 1, 1], 
                  index=["3", "3", "0", "4", "4", "2", "2", "5", "5", "0", "1", "6", "1", "6"])

>>> verts.reset_index().groupby('index').sum()
        0
index    
0      42
1       3
2      37
3      88
4      51
5      30
6       2

Or specify level=0 to group on the index.

verts.groupby(level=0).sum()