How to vectorize this operation on every row of a matrix

Question

I have a matrix filled with TRUE/FALSE values and I am trying to find the index position of the first TRUE value on each row (or retur

Answer 1

You can gain a lot of speed by using %% and %/%:

x <- matrix(rep(c(F,T,T),10), nrow=10)

z <- which(t(x))-1
((z%%ncol(x))+1)[match(1:nrow(x), (z%/%ncol(x))+1)]

This can be adapted as needed: if you want to do this for columns, you don't have to transpose the matrix.

Tried out on a 1,000,000 X 5 matrix :

x <- matrix(sample(c(F,T),5000000,replace=T), ncol=5)

system.time(apply(x,1,function(y) which(y)[1]))

#>   user  system elapsed 
#>  12.61    0.07   12.70 

system.time({
 z <- which(t(x))-1
 (z%%ncol(x)+1)[match(1:nrow(x), (z%/%ncol(x))+1)]}
)

#>   user  system elapsed 
#>   1.11    0.00    1.11 

You could gain quite a lot this way.