Melt a daframe with col names as values

Question

From a dataframe like this:

dframe = data.frame(data = c(1), error = c(3))

How can we melt the data aiming to have the col names as values.

Here an example of expected output

 dframe_ex = data.frame(id=c("data","error"), num=c(1,3))
 dframe_ex
     id num
1  data   1
2 error   3

Possible duplicate of [Reshape from wide to long format/structure](https://stackoverflow.com/questions/28261492/reshape-from-wide-to-long-format-structure) — NelsonGon, Feb 19 '19 at 01:10

score 2 · Answer 1 · edited Feb 19 '19 at 01:14

2

use this

library(tidyr)
dframe %>%
  gather(id, num)

output is

     id num
1  data   1
2 error   3

edited Feb 19 '19 at 01:14

NelsonGon

answered Feb 19 '19 at 01:07

Stephan

2

Or just `library(tidyr)`, unless you need other packages from `tidyverse`. – neilfws Feb 19 '19 at 01:08

score 2 · Accepted Answer · answered Feb 19 '19 at 01:08

2

This is a case for base R's stack

stack(dframe)
#  values   ind
#1      1  data
#2      3 error

answered Feb 19 '19 at 01:08

Maurits Evers

score 0 · Answer 3 · answered Feb 19 '19 at 11:51

0

you can use melt in reshape2 :

library(reshape2)
dframe_ex <- melt(dframe)

> print(dframe_ex)
  variable value
1     data     1
2    error     3

If you want to have the col names you specified then:

library(tidyverse)
dframe_ex <- melt(dframe) %>% 
  rename(id = variable, num = value)

answered Feb 19 '19 at 11:51

nycrefugee

The package `reshape2` has now been deprecated in favour of `tidyr` – Shinobi_Atobe Feb 19 '19 at 13:27
sure but since the OP did actually use the verb 'melt' and that the `reshape2` package still works, thought this would be of use. – nycrefugee Feb 19 '19 at 13:32

3 Answers3