Select columns by class (e.g. numeric) from a data.table

Question

Why doesn't this work with data.table?

It works with data.frame. Is there a way to do this with a data table?

x <- data.table(v1=1:20,v2=1:20,v3=1:20,v4=letters[1:20])
y <- x[ , sapply(x, is.numeric)]

This returns:

v1    v2    v3    v4
TRUE  TRUE  TRUE FALSE

Answer 1

From data.table 1.13.0 ".SDcols accepts a function which is used to select the columns of .SD". Thus, simply .SDcols = is.numeric:

x[ , .SD, .SDcols = is.numeric]

Answer 2

data.table needs the with=FALSE to grab column numbers.

tokeep <- which(sapply(x,is.numeric))
x[ , tokeep, with=FALSE]

Answer 3

You may also try:

 x1 <- x[,Filter(is.numeric, .SD)]
 head(x1,3)
 #   v1 v2 v3
#1:  1  1  1
#2:  2  2  2
#3:  3  3  3

Although, I have to admit that it is slow for bigger datasets.

Answer 4

Similar to @akrun's answer

Filter(is.numeric, x)

Answer 5

We can write a custom helper calledwhere(), and then we can subset a data.frame/data.table where f is satisfied:

where <- function(x, f) {
  colnames(x)[vapply(x, f, logical(1))]
}

df[, where(df, is.numeric), with = FALSE]