CODEWITHSUNDEEP
rvectorreplicate
Tracy
Tracy

Reputation: 709

Create new repeating index where range of each number is dependent on the index of another column as efficiently as possible

I have a vector of numbers.

initialindex= c(17,  23,  28,  34,  39,  45)

What I would like to get out of this looks like this:

finalindex=c(1,1,1,1,1,1,2,2,2,2,2,3,3,3,3,3,3,4,4,4,4,4,5,5,5,5,5,5,5)

The number repeats based on the difference of initialindex. 23-17= 6 1's and 28-23= 5'2s.

I can take the diff of initial index:

diff(initialindex)

which will give me the length of each value in the final index (6 1's, 5 2's, 6 3's). But, then I need to replicate them with the new index value 1: len(initialindex)

Can anyone help me with this?

Tracy

Upvotes: 4

Views: 164

Answers (3)

markus
markus

Reputation: 26343

A solution using rep – first proposed by @gung - Reinstate Monica but was deleted 

rep(x = 1:(length(initialindex) - 1L),
    times = diff(initialindex))
# [1] 1 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 5

Upvotes: 2

manotheshark
manotheshark

Reputation: 4357

Using inverse.rle

x <- rle(0)
x$lengths <- diff(initialindex)
x$values <- seq_along(x$lengths)
inverse.rle(x)

[1] 1 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 5

Upvotes: 1

tmfmnk
tmfmnk

Reputation: 39868

One option could be:

cumsum(sequence(diff(initialindex)) == 1)

 [1] 1 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 3 4 4 4 4 4 5 5 5 5 5 5

Upvotes: 3

Related Questions