CODEWITHSUNDEEP
rmatrixdataframeapply
Hardik Gupta
Hardik Gupta

Reputation: 4790

R: Convert Matrix to YES, NO dataframe

I have a matrix like this

df <- matrix(c(rep(1,3),rep(2,3)),nrow=3,ncol=2)

df: 
      [,1] [,2]
[1,]    1    2
[2,]    1    2
[3,]    1    2

I want to convert every cell value to YES, if greater than 0, else NO

I understand that I can do this using 

apply(df, 2, function(x) ifelse(x > 0, "Yes","No"))

However my matrix is very huge (million * 5000), and hence using apply takes insanely large time

I have also tried

df <- ifelse(df > 0, "Yes","No")

However even this takes a lot of time

Can I achieve better performance with this?

Upvotes: 2

Views: 800

Answers (2)

lmo
lmo

Reputation: 38500

If your goal is to visually examine the matrix, and not use the contents, theThe symnum function is designed for this use in particular. The Description section of ?symnum says

Symbolically encode a given numeric or logical vector or array. Particularly useful for visualization of structured matrices, e.g., correlation, sparse, or logical ones.

symnum(df, cutpoints=c(-Inf, 0, Inf), symbols=c("no", "yes"))

[1,] yes yes
[2,] yes yes
[3,] yes yes

Upvotes: 0

Sven Hohenstein
Sven Hohenstein

Reputation: 81683

Here's one way to create the matrix:

df[] <- c("No", "Yes")[(df > 0) + 1]

The result:

     [,1]  [,2] 
[1,] "Yes" "Yes"
[2,] "Yes" "Yes"
[3,] "Yes" "Yes"

Upvotes: 5

Related Questions