CODEWITHSUNDEEP
rduplicates
Gillette
Gillette

Reputation: 55

Remove repeated numbers in a sequence

I have a vector of the type 

c(3,3,...,9,9,...,2,2,...,3,3,...,7,7,...)

I want to remove the repeated numbers in a sequence, without breaking the order. This is, I would like to obtain something like 

c(3,9,2,3,7,...)

How can I do this in R?

Upvotes: 4

Views: 602

Answers (3)

RHertel
RHertel

Reputation: 23798

Just for the fun of it, here is an Rcpp version of solving the problem:

library(Rcpp)
cppFunction('NumericVector remove_multiples(NumericVector& vec) {   
   NumericVector c_vec(clone(vec));
   NumericVector::iterator it = std::unique(c_vec.begin(),c_vec.end());
   c_vec.erase(it,c_vec.end());
   return(c_vec);
  }'
)

x <- c(1,1,1,2,2,2,1,1,3,4,4,1,1)    
> remove_multiples(x)
[1] 1 2 1 3 4 1

Upvotes: 2

Heroka
Heroka

Reputation: 13149

We can also use the observation that a duplicate in sequence has a difference of 0 with their neighbour. Therefore, using base-R, we can do:

v[c(1,diff(v))!=0]

Upvotes: 9

akrun
akrun

Reputation: 887251

We can try with rleid and duplicated. We create the run-length-ids with rleid (from data.table) so that only adjacent elements that are equal forms one group, get the logical index of not duplicated values and subset the vector.

library(data.table)
v1[!duplicated(rleid(v1))]
#[1] 3 9 2 3 7

Or as the OP mentioned, we can use rle from base R and extract the values.

rle(v1)$values
#[1] 3 9 2 3 7

data

 v1 <- c(3,3,9,9,2,2,3,3,7,7)

Upvotes: 7

Related Questions