2

I have two matrices:For example

temp1 <- matrix(c(1,2,3,4,5,6),2,3,byrow = T)
temp2 <- matrix(c(7,8,9),1,3,byrow = T)

temp1

       [,1] [,2] [,3]
 [1,]    1    2    3
 [2,]    4    5    6

temp2

       [,1] [,2] [,3]
 [1,]    7    8    9

I have two matrices with the same number of rows, but with different rows. I would like to add these two matrices as follows. I wonder if there is a way to add R without for statements and apply functions.

temp <- do.call(rbind,lapply(1:2,function(x){temp[x,]+temp2}))

temp

       [,1] [,2] [,3]
 [1,]    8   10   12
 [2,]   11   13   15

This example is simple, but in practice I need to do the above with a 100 * 100 matrix and a 1 * 100 matrix. In this case, it takes too long, so I do not want to use for statements and apply functions.

Sunsick Choo
  • 23
  • 3
  • You can transpose the 100x100 matrix, using `t`, then add the 1*100 matrix with `as.numeric`, then transpose the result. i.e `t(t(temp1) + as.numeric(temp2))` – Aaron Hayman Sep 27 '18 at 10:33
  • 2
    Alternatively, you can expand the later one to the same shape as the former: `temp1 + temp2[rep(1, nrow(temp1)), ]` – mt1022 Sep 27 '18 at 10:53

1 Answers1

2

You can use ?sweep:

temp1 <- matrix(c(1,2,3,4,5,6),2,3,byrow = T)
temp2 <- matrix(c(7,8,9),1,3,byrow = T)
sweep(temp1, 2, temp2, '+')

Unfortunately the help for sweep is really difficult to understand, but in this example you apply the function ´+´ with argument ´temp2´ along the second dimension of temp1.

For more examples, see: How to use the 'sweep' function

Otto Kässi
  • 2,943
  • 1
  • 10
  • 27