CODEWITHSUNDEEP
rduplicates
Jennifer Therese
Jennifer Therese

Reputation: 1295

How to identify columns in a data frame that has same value?

I have a data frame

         A B C D E F
         1 2 3 3 4 1
         2 3 5 5 8 2
         6 4 9 9 0 6
         1 2 3 3 8 1

I want the names of columns that holds identical values.

Preferred output : A,F,C,D

Upvotes: 2

Views: 332

Answers (3)

s_baldur
s_baldur

Reputation: 33603

You could convert the data.frame to a list and use duplicated():

names(df)[duplicated(as.list(df)) | duplicated(as.list(df), fromLast = TRUE)]
# [1] "A" "C" "D" "F"

You could also duplicated.default() directly on the data.frame:

names(df)[duplicated.default(df) | duplicated.default(df, fromLast = TRUE)]
# [1] "A" "C" "D" "F"

Data:

df <- data.frame(
  A = c(1L, 2L, 6L, 1L), B = c(2L, 3L, 4L, 2L), C = c(3L, 5L, 9L, 3L),
  D = c(3L, 5L, 9L, 3L), E = c(4L, 8L, 0L, 8L), F = c(1L, 2L, 6L, 1L)
)

Benchmark:

Converting a data.frame to a list (as.list()) is much more efficient than transposing and converting to a matrix (t()):

microbenchmark::microbenchmark(as.list(df), t(df))
Unit: microseconds
        expr    min      lq     mean median      uq     max neval cld
 as.list(df)  2.677  2.9010  3.84244  3.570  3.5700  28.114   100  a 
       t(df) 69.615 71.1765 77.11636 72.293 75.6395 219.554   100   b

Upvotes: 1

symbolrush
symbolrush

Reputation: 7467

Expanding @Ronak Shah's solution to produce OPs preferred output:

df <- data.frame(A = c(1,2,6,1), B = c(2,3,4,2), C = c(3,5,9,3), D = c(3,5,9,3), E = c(4,8,0,8), F = c(1,2,6,1))
df <- df[, duplicated(t(df)) | duplicated(t(df), fromLast = TRUE)]
df <- df[order(df[1])]
names(df)
[1] "A" "F" "C" "D"

Note that order() is used so rearrange df so that names(df) returns the preferred output.

Upvotes: 2

Ronak Shah
Ronak Shah

Reputation: 389235

duplicated works on each row of data frame. We could transpose it to make it work for each column and subset the names of the columns.

names(df)[duplicated(t(df)) | duplicated(t(df), fromLast = TRUE)]
#[1] "A" "C" "D" "F"

Upvotes: 1

Related Questions