Covert all column headers to a single column in dataframe in Python

Question

I have a dataframe looks like below:

   date       A       B       C
2015-01-25  0.97    0.33    0.65
2015-02-01  0.23    0.02    0.67

My goal is to covert the df to a new format like below:

  date     Type Value
2015-01-25  A   0.97
2015-02-01  A   0.23
2015-01-25  B   0.33
2015-02-01  B   0.08
2015-01-25  C   0.65
2015-02-01  C   0.67

which means move all the columns to the rows as a new column as type, and put their value for each date as a unique columns, I did some search on Google, but still cannot find a right solution.

Can anyone please help?

Jaroslav Bezděk · Accepted Answer · 2020-04-16T06:17:37.753

1

You can use pandas.melt() method like this:

>>> df = df.melt(id_vars=['date'], var_name='Type', value_name='Value')
>>> print(df)
         date Type  Value
0  2015-01-25    A   0.97
1  2015-02-01    A   0.23
2  2015-01-25    B   0.33
3  2015-02-01    B   0.02
4  2015-01-25    C   0.65
5  2015-02-01    C   0.67

edited Apr 16 '20 at 06:17

answered Apr 16 '20 at 06:13

Jaroslav Bezděk

6,967
6
29
46

Ok, I am sorry. – Jaroslav Bezděk Apr 16 '20 at 06:16
Ya, it is up to you if remove answer or not ;) Obviously if youare rep hunter not removed ;) – jezrael Apr 16 '20 at 06:18

score 1 · Answer 2 · answered Apr 16 '20 at 06:19

1

Use melt()

df.melt(id_vars='date', var_name='Type', value_name='Values')

answered Apr 16 '20 at 06:19

Peter

10,959
2
30
47

Covert all column headers to a single column in dataframe in Python

2 Answers2