How do I get unique element from a vector, keeping its name?

Question

I know there is the function unique() that extracts the unique values from a vector. But I lose its name.

Ex.

vector = c("A" = 1, "B" = 2, "A" = 1, "C" = 3, "B" = 2, "D" = 3, "D" = 3)

If I print I should see:

A B A C B D D
1 2 1 3 2 3 3

Expected output:

A B C D
1 2 3 3

Attempts:

If I use: unique(vector) I only get 1 2 3

If I use: vector[!duplicated(vector)] I get:

A B C 
1 2 3

This is close, but the "D" = 3 is missing.

`x[!duplicated(paste(x, names(x)))]`? – talat Mar 10 '17 at 09:22 — talat, Mar 10 '17 at 09:22

Ronak Shah · Answer 1 · 2017-03-10T09:43:55.233

6

vector = c(A=1,B=2,A=1,C=3,B=2,D=3,D=3)

When you do,

vector[!duplicated(vector)]

it looks for duplicates in values of vector and not names hence the output which you get is

A B C 
1 2 3

If you want to find unique names then you should run duplicated function on the names of the vector

vector[!duplicated(names(vector))]

A B C D 
1 2 3 3

Also similar ouptut could be achieved using unique

vector[unique(names(vector))]

A B C D 
1 2 3 3

edited Mar 10 '17 at 09:43

answered Mar 10 '17 at 09:04

Ronak Shah

377,200
20
156
213

I think they want unique on the values, try: `vector = c(A=1,B=2,A=1,C=3,B=2,D=3,D=4)` – zx8754 Mar 10 '17 at 09:06
Great! I just saw I made a mistake on the vector, but yes this is what I needed! – Saul Garcia Mar 10 '17 at 09:07
I'd simply do this : `v[!duplicated(names(v)) | !duplicated(v)]` – digEmAll Mar 10 '17 at 09:07
@SaulGarcia this is not what you want, try with D=3 and D=4 instead of D=3, D=3. – zx8754 Mar 10 '17 at 09:08
@zx8754 If I have understood OP correctly, they just need unique names and not unique names and unique values. – Ronak Shah Mar 10 '17 at 09:10
If that is the case, why not just use `unique(names(v))` ? – zx8754 Mar 10 '17 at 09:10
@zx8754 because OP needs the values of those names as well. – Ronak Shah Mar 10 '17 at 09:14
So if it is "D=3,D=4", then we just pick the first one? – zx8754 Mar 10 '17 at 09:15
Atleast `duplicated` does that. If OP has different requirement they can clarify. – Ronak Shah Mar 10 '17 at 09:16

akrun · Answer 2 · 2017-03-10T09:40:41.520

2

We can use match

vector[match(unique(names(vector)), names(vector))]
# A B C D 
# 1 2 3 3

or with tapply

tapply(vector, names(vector), FUN = head, 1)
#  A B C D 
# 1 2 3 3

Or using data.table

library(data.table)
data.table(Key = names(vector), Value = c(vector))[, Value[1L], Key]

edited Mar 10 '17 at 09:40

answered Mar 10 '17 at 09:25

akrun

874,273
37
540
662

score 2 · Answer 3 · answered Mar 10 '17 at 09:33

2

Just to add another alternative that also may cover discripancy between values and names

library(dplyr)
data_frame(value = v, name = names(v)) %>%
  group_by(name, value) %>% # alternatively: group_by(name) if name value pair is always unique
  slice(1) %>%
  ungroup()

answered Mar 10 '17 at 09:33

Drey

3,314
2
21
26

Excellent solution, just that `group_by` can be slow when the df is large. – Conner M. Feb 10 '20 at 16:39

How do I get unique element from a vector, keeping its name?

3 Answers3