interpolation with matrices in R

问题

I want to use linear interpolation to create a large matrix from 2 smaller matrices. I can do this using a function like this:

mat1 <- matrix(rep(20, 4), ncol = 2)
mat2 <- matrix(seq(21, 24, 1), ncol = 2)
mat3 <- matrix(c(18, 27, 25, 12), ncol = 2)

num.days <- c(31, 29)

interpolate <- function(initial, final, n){
  data.list <- list()
  for (i in 1:(n - 1)){
    step1 <- (final - initial) / n
    step2 <- step1 * i
    data.list[[1]] <- initial
    data.list[[i+1]] <- round(step2 + initial, 2)
  }
  newmat = do.call(cbind, data.list)
  return(newmat)
}


interpolate(mat1, mat2, num.days[1])
interpolate(mat2, mat3, num.days[2])

I want to modify this code so that this function is performed iteratively over many matrices. I've tried putting the matrices in a list and rewriting the function to interpolate between each matrix within the list but haven't been able to get it to work. Any help or suggestions are greatly appreciated. Thanks.

回答1:

You could just put all objects into a list

l <- list(mat1, mat2, mat3, num.days)

and use lapply()

lapply(1:length(l[[4]]), function(x) interpolate(l[[x]], l[[x + 1]], l[[4]][x]))

Yielding

List of 2
 $ : num [1:2, 1:62] 20 20 20 20 20 ...
 $ : num [1:2, 1:58] 21 22 23 24 20.9 ...

来源：https://stackoverflow.com/questions/51597482/interpolation-with-matrices-in-r