[R] table over a matrix dimension...

[Marc Schwartz]

On Jul 10, 2014, at 12:03 PM, Jonathan Greenberg <jgrn at illinois.edu> wrote:

> R-helpers:
> 
> I'm trying to determine the frequency of characters for a matrix
> applied to a single dimension, and generate a matrix as an output.
> I've come up with a solution, but it appears inelegant -- I was
> wondering if there is an easier way to accomplish this task:
> 
> # Create a matrix of "factors" (characters):
> random_characters=matrix(sample(letters[1:4],1000,replace=TRUE),100,10)
> 
> # Applying with the table() function doesn't work properly, because not all rows
> # have ALL of the factors, so I get a list output:
> apply(random_characters,1,table)
> 
> # Hacked solution:
> unique_values = letters[1:4]
> 
> countsmatrix <- t(apply(random_characters,1,function(x,unique_values)
> {
> counts=vector(length=length(unique_values))
> for(i in seq(unique_values))
> {
> counts[i] = sum(x==unique_values[i])
> }
> return(counts)
> },
> unique_values=unique_values
> ))
> 
> # Gets me the output I want but requires two nested loops (apply and
> for() ), so
> # not efficient for very large datasets.
> 
> ###
> 
> Is there a more elegant solution to this?
> 
> --j
> 


If I am correctly understanding your issue, you simply need to coerce the input to table() to a factor with a common set of levels, since the matrix will be 'character' by default:


set.seed(1)
random_characters <- matrix(sample(factor(letters[1:4]), 1000, replace = TRUE), 100, 10)

> random_characters 
       [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10]
  [1,] "b"  "c"  "b"  "c"  "c"  "c"  "d"  "d"  "d"  "d"  
  [2,] "b"  "b"  "a"  "a"  "a"  "c"  "d"  "d"  "a"  "d"  
  [3,] "c"  "b"  "c"  "b"  "d"  "c"  "a"  "d"  "d"  "b"  
  [4,] "d"  "d"  "b"  "b"  "d"  "c"  "c"  "c"  "c"  "a"  
  [5,] "a"  "c"  "a"  "b"  "d"  "b"  "d"  "c"  "b"  "a"  
  [6,] "d"  "a"  "c"  "d"  "c"  "d"  "d"  "a"  "c"  "a"  
  [7,] "d"  "a"  "c"  "a"  "b"  "b"  "b"  "b"  "b"  "a"  
  [8,] "c"  "b"  "a"  "d"  "d"  "d"  "b"  "c"  "d"  "a"  
  [9,] "c"  "d"  "b"  "a"  "a"  "d"  "d"  "d"  "b"  "a"  
 [10,] "a"  "c"  "c"  "b"  "d"  "c"  "a"  "c"  "a"  "a"  
 [11,] "a"  "d"  "d"  "a"  "d"  "d"  "d"  "c"  "b"  "c"  
 [12,] "a"  "c"  "a"  "a"  "b"  "b"  "b"  "b"  "b"  "d"  
 [13,] "c"  "b"  "d"  "d"  "c"  "a"  "c"  "a"  "b"  "c"  
 [14,] "b"  "b"  "d"  "c"  "d"  "c"  "c"  "d"  "d"  "a"  
 [15,] "d"  "a"  "d"  "b"  "c"  "c"  "c"  "b"  "b"  "a"  
 [16,] "b"  "a"  "b"  "b"  "b"  "a"  "b"  "b"  "c"  "b"  
 [17,] "c"  "c"  "c"  "a"  "b"  "c"  "a"  "a"  "d"  "a"  
 [18,] "d"  "a"  "d"  "b"  "b"  "c"  "b"  "a"  "d"  "c" 
 ...


RES <- t(apply(random_characters, 1, 
               function(x) table(factor(x, levels = letters[1:4]))))

> RES
       a b c d
  [1,] 0 2 4 4
  [2,] 4 2 1 3
  [3,] 1 3 3 3
  [4,] 1 2 4 3
  [5,] 3 3 2 2
  [6,] 3 0 3 4
  [7,] 3 5 1 1
  [8,] 2 2 2 4
  [9,] 3 2 1 4
 [10,] 4 1 4 1
 [11,] 2 1 2 5
 [12,] 3 5 1 1
 [13,] 2 2 4 2
 [14,] 1 2 3 4
 [15,] 2 3 3 2
 [16,] 2 7 1 0
 [17,] 4 1 4 1
 [18,] 2 3 2 3
 ...



Regards,

Marc Schwartz