How to melt 2 columns at the same time?

In Pandas, I have the following data frame:

   id1 id2 t1  l1  t2  l2 
0  1   2   a   b   c   d
1  3   4   g   h   i   j

I would like to melt two columns at once. That is, the desired output is:

   id1 id2 tz  lz  
0  1   2   a   b
1  1   2   c   d
2  3   4   g   h
3  3   4   i   j

I know standard melting:

d.melt(id_vars=['id1', 'id2'],
       value_vars=['t1', 't2', 'l1', 'l2'])

but that stacks all columns

   id1  id2 variable value
0    1    2       t1     a
1    3    4       t1     g
2    1    2       t2     c
3    3    4       t2     i
4    1    2       l1     b
5    3    4       l1     h
6    1    2       l2     d
7    3    4       l2     j

How could I melt two columns at once? Something like:

d.melt(id_vars=['id1', 'id2'],
       value_vars={('t1', 'l1'): 'tz', ('t2', 'l2'): 'lz'})

would be great.

Answers:

Thank you for visiting the Q&A section on Magenaut. Please note that all the answers may not help you solve the issue immediately. So please treat them as advisements. If you found the post helpful (or not), leave a comment & I’ll get back to you as soon as possible.

Method 1

This is wide_to_long

pd.wide_to_long(df,['t','l'],i=['id1','id2'],j='drop').reset_index(level=[0,1])
Out[52]: 
      id1  id2  t  l
drop                
1       1    2  a  b
2       1    2  c  d
1       3    4  g  h
2       3    4  i  j

Method 2

You can use melt twice here and after that concat them to get desired output:

t = d.melt(id_vars=['id1', 'id2'], value_vars=['t1', 't2'], value_name='tz').drop('variable', axis=1)
l = d.melt(id_vars=['id1', 'id2'], value_vars=['l1', 'l2'], value_name='lz').iloc[:, -1:]

df = pd.concat([t, l], axis=1).sort_values('id1')

Output

print(df)
   id1  id2 tz lz
0    1    2  a  b
2    1    2  c  d
1    3    4  g  h
3    3    4  i  j

All methods was sourced from stackoverflow.com or stackexchange.com, is licensed under cc by-sa 2.5, cc by-sa 3.0 and cc by-sa 4.0