Transposing rows to single column by group pandas

Question

Can someone help me with this transformation. I have a dataset:

ID	a1	a2	a3
a	1	5	3
a	2	3	3
b	2	3	5
b	3	5	6
c	5	1	3
c	2	4	2

and i want it like this:

ID	Levels
a	1
a	5
a	3
a	2
a	3
a	3
b	2
b	3
b	5
b	3
b	5
b	6

..... and so forth.Thanks!

Answer 1

Use a 'melt' function:

df_melted = pd.melt(df, id_vars=['ID'], value_vars=['a1', 'a2', 'a3'], var_name='Levels', value_name='Values')

Answer 2

A possible solution:

df.set_index('ID').stack().droplevel(1).reset_index(name='Levels')

Output:

   ID  Levels
0   a       1
1   a       5
2   a       3
3   a       2
4   a       3
5   a       3
6   b       2
7   b       3
8   b       5
9   b       3
10  b       5
11  b       6
12  c       5
13  c       1
14  c       3
15  c       2
16  c       4
17  c       2

Answer 3

You can use pd.MultiIndex

names = ["ID", "Levels"]
data = <your data>

index = pd.MultiIndex.from_product([range(i) for i in data.shape], names=names)

df = pd.DataFrame(data, index=index)