CODEWITHSUNDEEP
rrounding
Tajda
Tajda

Reputation: 29

Rounding off all numeric columns in data frame to different numbers of digits

I have a data frame containing many columns of different classes, I'm including a very simplified example below:

df <- data.frame( col1=c("mary", "john", "luke", "peter", "agnes", "anne", "july", "judy", "tina", "ben"),
                  col2=c(0.1111,0.3123,0.7875,0.6789,0.87862,0.37512,0.69875,0.1785,0.1777, 0.9999),
                  col3=c(0.1111,0.3123,0.7875,0.6789,0.87862,0.37512,0.69875,0.1785,0.1777, 0.9999),
                  col3=c(124.6, 763.8, 222.9, 765.0, 564.7, 761.8, 683.9, 655.9, 897.3, 566.1))

I would like to round off all of the numeric columns, but to different values. For example: col2 and col3 should be rounded off to 2 digits and col4 to 0 digits.

The style of code I was able to come up with is something like this:

df2 <- df %>% mutate_at(2, round, 2)
df3 <- df2 %>% mutate_at(3, round, 2) 
df4 <- df3 %>% mutate_at(4, round, 0)

If I don't assign the value to a new data frame each time around it doesn't save the values to my data frame. For instance if I do something like this:

df %>% mutate_at(2, round, 2)
df %>% mutate_at(3, round, 2) 
df %>% mutate_at(4, round, 0)

It writes out the desired output to the console, but doesn't save it to df. I was wondering if there was more elegant and intuitive way of doing this?

Upvotes: 0

Views: 360

Answers (2)

Ronak Shah
Ronak Shah

Reputation: 388907

_at, _all, _if variants are now superseded by across which also allows you to input multiple operations in the same mutate pipe.

library(dplyr)

df <- df %>%
        mutate(across(2:3, round, 2), 
               across(4, round, 0))
df

#    col1 col2 col3 col3.1
#1   mary 0.11 0.11    125
#2   john 0.31 0.31    764
#3   luke 0.79 0.79    223
#4  peter 0.68 0.68    765
#5  agnes 0.88 0.88    565
#6   anne 0.38 0.38    762
#7   july 0.70 0.70    684
#8   judy 0.18 0.18    656
#9   tina 0.18 0.18    897
#10   ben 1.00 1.00    566

Upvotes: 1

tonybot
tonybot

Reputation: 655

You can keep using the pipe operator to connect the code together.

df <- df %>% 
  mutate_at(c(2, 3), round, 2) %>%
  mutate_at(4, round, 0)

Upvotes: 2

Related Questions