How to use apply function in a pipe operator

问题

I have a dataframe which has a few character columns followed by a few numerical columns. I want to add a new column using the %>% operators which is the highest value from the numerical columns per row.

let's say the data frame looks like this:

character1, character2, value1, value2, value3

"string", "string", 5, 7, 4

"string", "string", 3, 4, 2

"string", "string", 2, 8, 6

Then the new column should be 7 for the first row, 4 for second row and 8 for last row. I am trying to use the apply function in the pipe operator but it's not working properly.

new_df <- old_df %>%
  mutate(new_column = apply(value1:value3, 1, max))

It returns error: Numerical expression has XXX values only first used.

I also tried using max(value1: value3) instead of apply but that doesn't work either.

回答1:

You can use rowwise from dplyr:

library(dplyr)
df %>% 
  rowwise() %>% 
  mutate(new_column = max(c_across(value1:value2)))

# A tibble: 3 x 6
# Rowwise: 
  character1 character2 value1 value2 valu3 new_column
  <chr>      <chr>       <int>  <int> <int>      <int>
1 string     string          5      7     4          7
2 string     string          3      4     2          4
3 string     string          2      8     6          8

Data

library(tidyverse)
df <- tibble::tribble(
          ~character1,           ~character2, ~value1, ~value2, ~valu3,
  "string", "string", 5L, 7L, 4L,
  "string", "string", 3L, 4L, 2L,
  "string", "string", 2L, 8L, 6L
  )

回答2:

We can also use pmax

library(dplyr)
library(purrr)
df %>%
   mutate(new_column = reduce(select(., starts_with('value')), pmax))
# A tibble: 3 x 6
#  character1 character2 value1 value2 valu3 new_column
#  <chr>      <chr>       <int>  <int> <int>      <int>
#1 string     string          5      7     4          7
#2 string     string          3      4     2          4
#3 string     string          2      8     6          8

data

df <- structure(list(character1 = c("string", "string", "string"), 
    character2 = c("string", "string", "string"), value1 = c(5L, 
    3L, 2L), value2 = c(7L, 4L, 8L), valu3 = c(4L, 2L, 6L)), row.names = c(NA, 
-3L), class = c("tbl_df", "tbl", "data.frame"))

来源：https://stackoverflow.com/questions/62247801/how-to-use-apply-function-in-a-pipe-operator

标签

dplyr