CODEWITHSUNDEEP
rlistmatrixlapplymatrix-multiplication
asm
asm

Reputation: 692

Use lapply to multiply two lists of matrices in R

I have two lists, each one with many matrices. For brevity, here is an example with two matrices per list:

l1 <- list(matrix(1:20, nrow=2),
           matrix(21:40, nrow=2))

l2 <- list(matrix(41:60, nrow=2),
           matrix(61:80, nrow=2))

I am trying to use lapply to (a) remove the first column of each matrix and then (b) multiply each matrix in l1 against the corresponding matrix in l2.

First, I tried:

result <- lapply(l1, function(x,y){
  
  tmp <- x[,-1]*y[,-1]
  return(tmp)
  
}, y=l2)

It didn't work. The problem is that I can't remove the first column from l2 this way. If l2 were a vector instead of a list, the code would work with tmp <- x[,-1]*y[-1]. But even when I try to simply multiply the matrices in each list, it still does not work:

result <- lapply(l1, function(x,y){
  
  tmp <- x*y
  return(tmp)
  
}, y=l2)

Any ideas?

Upvotes: 1

Views: 600

Answers (2)

akrun
akrun

Reputation: 887048

Using map2

library(purrr)
map2(l1, l2, ~ .x[, -1] * .y[,-1])
[[1]]
     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
[1,]  129  225  329  441  561  689  825  969 1121
[2,]  176  276  384  500  624  756  896 1044 1200

[[2]]
     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
[1,] 1449 1625 1809 2001 2201 2409 2625 2849 3081
[2,] 1536 1716 1904 2100 2304 2516 2736 2964 3200

Upvotes: 2

Ronak Shah
Ronak Shah

Reputation: 388907

In lapply, you can iterate over only one object. You may use Map here -

Map(function(x, y)  x[,-1]*y[,-1], l1, l2)

#[[1]]
#     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
#[1,]  129  225  329  441  561  689  825  969 1121
#[2,]  176  276  384  500  624  756  896 1044 1200

#[[2]]
#     [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9]
#[1,] 1449 1625 1809 2001 2201 2409 2625 2849 3081
#[2,] 1536 1716 1904 2100 2304 2516 2736 2964 3200

To use lapply you may use the index to subset both l1 and l2.

lapply(seq_along(l1), function(x) l1[[x]][, -1] * l2[[x]][, -1])

Upvotes: 2

Related Questions