CODEWITHSUNDEEP
rdata.table
ATMathew
ATMathew

Reputation: 33

How to use ifelse across multiple columns in a data.table?

I have multiple columns in a data.table that I would like to apply an ifelse statement on. I could repeat the code for every column, but am wondering if there is a more elegant solution

For a given data table, I want to alter some numeric columns using the ifelse statement.

Here is the data:

dt = data.table(id = c(101, 102, 103, 104), one_cnt = c(0, 6, 4, 0), two_cnt = c(4, 0, 0, 2))

dt

If I wanted to do this manually, I could do this across all columns

dt$one_cnt = ifelse(dt$one_cnt == 0, 0, 2)

Instead, I'd like to select the following columns and apply the ifelse on each

colnames(dt)[grepl("cnt", colnames(dt))]

for one_cnt and two_cnt, a 0 value would end up with a 0. A non-zero value would take on a 1 value

Upvotes: 2

Views: 1245

Answers (1)

Ronak Shah
Ronak Shah

Reputation: 389325

We could use .SDcols to specify selected columns. 

library(data.table)
cols <- grep("cnt", colnames(dt), value = TRUE)
dt[, (cols) :=  lapply(.SD, function(x) ifelse(x == 0, 0, 2)), .SDcols = cols]

dt
#    id one_cnt two_cnt
#1: 101       0       2
#2: 102       2       0
#3: 103       2       0
#4: 104       0       2

From your attempt you are replacing 0 values with 0 and changing non-zero value to 2 but in your description you mentioned that you want to change the non-zero value to 1, so I am not clear about your expected output. If you want to change it to 1, you can do

dt[, (cols) :=  lapply(.SD, function(x) as.integer(as.logical(x))), .SDcols = cols]

Upvotes: 4

Related Questions