Mathematical operations on columns specified by vector in R

Question

I need to replace specific values in a subset of columns in my dataframe. Specifically, I need to replace the value 1 in this subset of columns with 0.9999. I created a vector containing the column names of columns where replacement is needed. These columns must be subset by name and not column number.

peaches <- c( 0, 1, 0, 1)
bananas <- c( 0, 1, 1, 1)
apples <- c( 1, 1, 1, 1)
oranges <- c (0, 0, 0, 1)
fruits <- data.frame(peaches, bananas, apples, oranges)
vector <- c("apples", "bananas", "peaches")

My first attempt looked like this:

fruits[vector][fruits[vector] == 1] <- 0.9999

While it works for this small dataset, it is not transforming values in the large dataset I'm working with. Anyone know why?

Answer 1

You can try :

fruits[,vector] <- setDT(lapply(fruits[,vector], function(x) replace(x, x==1, 0.9999)))

  peaches bananas apples oranges
1  0.0000  0.0000 0.9999  0.0000
2  0.9999  0.9999 0.9999  0.0000
3  0.0000  0.9999 0.9999  0.0000
4  0.9999  0.9999 0.9999  0.9999

Answer 2

Tidyverse solution

library(tidyverse)

peaches <- c( 0, 1, 0, 1)
bananas <- c( 0, 1, 1, 1)
apples <- c( 1, 1, 1, 1)
oranges <- c (0, 0, 0, 1)
fruits <- data.frame(peaches, bananas, apples, oranges)
vector <- c("apples", "bananas", "peaches")

fruits %>% 
  mutate_at(vector, ~ replace(., . == 1, .9999))
#>   peaches bananas apples oranges
#> 1  0.0000  0.0000 0.9999       0
#> 2  0.9999  0.9999 0.9999       0
#> 3  0.0000  0.9999 0.9999       0
#> 4  0.9999  0.9999 0.9999       1

^{Created on 2020-01-14 by the reprex package (v0.3.0)}

Answer 3

Try this :

fruits[,vector][fruits[,vector] == 1] <- 0.9999

fruits[vector] uses the fact that a data.frame is a list of vector (of the same length), maybe the "matrix" approach is more adapted.