Pandas converting column to rows [duplicate]

Question

I have the following Python pandas dataframe:

id| country  | 2016 | 2017 | 2018
--+----------+------+------+------
0 | saudi    | A    | null | B
1 | china    | C    | A    | B
2 | pakistan | null | null | C

I want:

id| country  | year | value
--+----------+------+------
0 | saudi    | 2016 | A
1 | saudi    | 2018 | B
2 | china    | 2016 | C
3 | china    | 2017 | A
4 | china    | 2018 | B
5 | pakistan | 2018 | C

Answer 1

You can use df.melt

df.melt(['id','country'], df.columns[2:]).dropna()

    id  country variable  value
0   0   saudi     2016     A
6   0   saudi     2018     B
1   1   china     2016     C
4   1   china     2017     A
7   1   china     2018     B
8   2   pakistan  2018     C

Then just add .sort_values('id') if you want

Or you can use stack

df.set_index(['id','country']).stack().to_frame()

and then .reset_index() if you want

It looks like using stack will be a little more efficient:

melt: 11.3 ms ± 798 µs per loop (mean ± std. dev. of 3 runs, 1000 loops each)

stack: 9.18 ms ± 594 µs per loop (mean ± std. dev. of 3 runs, 1000 loops each)

Answer 2

yes you may also use unstack df2=df.unstack() and then df2.reset_index()