Scale numbers based on the largest value-CodePudding

I have data like this:

mydata <- c(365673, 478934, 897464, 254778, 678899)

I want to transform mydata using unique number. This is how I calculate the unique number:

Extract the largest number, in this case, it is 897464
After that we get the unique number is 800000 from 897464
Then divide the vector with that value

The output that I want is going to be:

newdata <- mydata / 800000
newdata
# [1] 0.4570913 0.5986675 1.1218300 0.3184725 0.8486238

How do I create the function?

CodePudding user response：

I think this is what you are trying to do:

scale.nums <- function(v){
  x = max(v)
  pow <- 10^floor(log10(x))
  x <- x %/% pow * pow
  v/x
}

For example:

> mydata <- c(365673, 478934, 897464, 254778, 678899)
> scale.nums(mydata)
[1] 0.4570913 0.5986675 1.1218300 0.3184725 0.8486238

CodePudding user response：

plyr has round_any function which would be useful here.

mydata <- c(365673, 478934, 897464, 254778, 678899)

mydata/max(plyr::round_any(mydata, 100000, floor))
#[1] 0.4570913 0.5986675 1.1218300 0.3184725 0.8486238