Align rows of dataframes with different sizes

Question

I have multiple DataFrames and they all have different sizes:

a

	Header1	Header2
111	item1	item2
112	item3	item4

a = pd.DataFrame({
  'Header1': ['item1', 'item3'],
  'Header2': ['item2', 'item4']
}).set_index(pd.Index([111, 112]))

b

	Header3	Header4
111	item5	item6
111	item7	item8
111	item9	item10
112	item11	item12
112	item13	item14
112	item15	item16

b = pd.DataFrame({
  'Header3': ['item5', 'item7', 'item9', 'item11', 'item13', 'item15'],
  'Header4': ['item6', 'item8', 'item10', 'item12', 'item14', 'item16']
}).set_index(pd.Index([111, 111, 111, 112,  112, 112]))

I would like to concatenate these two DataFrames like this:

c

	Header1	Header2	Header3	Header4
111	item1	item2	item5	item6
111	-	-	item7	item8
111	-	-	item9	item10
112	item3	item4	item11	item12
112	-	-	item13	item14
112	-	-	item15	item16

When I add rows into the smaller DataFrames I still don't know how to line up the rows like this.

score 1 · Answer 1 · answered Dec 15 '21 at 00:37

You can simply use pd.concat, which automatically takes indexes into account:

c = pd.concat([a, b], axis=1)

Output:

>>> c
    Header1 Header2 Header3 Header4
111   item1   item2   item5   item6
111   item1   item2   item7   item8
111   item1   item2   item9  item10
112   item3   item4  item11  item12
112   item3   item4  item13  item14
112   item3   item4  item15  item16

score 1 · Answer 2 · answered Dec 15 '21 at 00:49

Assuming the dashes in Header1 and Header2 represent filling the same values, it's simplest to join a and b:

c = a.join(b)

#     Header1 Header2 Header3 Header4
# 111   item1   item2   item5   item6
# 111   item1   item2   item7   item8
# 111   item1   item2   item9  item10
# 112   item3   item4  item11  item12
# 112   item3   item4  item13  item14
# 112   item3   item4  item15  item16

Align rows of dataframes with different sizes

a

b

c

2 Answers2