CODEWITHSUNDEEP
r
R J
R J

Reputation: 1

Search for TRUE in a vector containing TRUE / FALSE

A<-(FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE)

I want an efficient way to find if A has TRUE and stop when TRUE is found.

Upvotes: 0

Views: 138

Answers (3)

PavoDive
PavoDive

Reputation: 6496

My first reaction was thinking about sum(A) != 0 but then went for the microbenchmark and these are the results:

# install.packages("microbenchmark")
require(microbenchmark)
microbenchmark("sum" = {A <- sample(c(TRUE, FALSE), 1e6, TRUE); sum(A) != 0}, "any" = {A <- sample(c(TRUE, FALSE), 1e6, TRUE); any(A)}, "seq_len" = {A <- sample(c(TRUE, FALSE), 1e6, TRUE); A[seq_len(which(A)[1])]}, times = 100L)

Unit: milliseconds
    expr      min       lq      mean   median       uq      max neval
     sum 72.98672 75.10716  86.33333 76.88114 78.57807 216.4087   100
     any 68.59519 72.35810  78.29730 72.50796 72.83563 229.6932   100
 seq_len 80.41055 84.41244 104.30460 87.82726 90.15668 223.8297   100

So definitely any seems to be the most efficient (other than a Rcpp implementation)

Upvotes: 0

Stewart Macdonald
Stewart Macdonald

Reputation: 2132

If you just want to know if A contains any TRUE values, you can use any():

> A <- c(FALSE, FALSE, FALSE, FALSE, FALSE, TRUE, FALSE, FALSE, FALSE)
> any(A)
[1] TRUE

Upvotes: 2

akrun
akrun

Reputation: 887058

We can use which to find the elements that are TRUE, select the first, sequence it and subset the 'A'

A[seq_len(which(A)[1])]

data

A <- c(FALSE, FALSE, FALSE, FALSE, FALSE, TRUE, FALSE, FALSE, FALSE)

Upvotes: 1

Related Questions