根据 R dplyr 中的条件将字符转换为 NA

Question

我有一个看起来像这样的数据框

library(tidyverse)
df3 <- tibble(col1 = c("apple",rep("banana",3)),
              col2 = c("aple", "banan","bananb","banat"), 
              count_col1 = c(1,4,4,4), 
              count_col2 = c(4,1,1,1))
df3
#> # A tibble: 4 × 4
#>   col1   col2   count_col1 count_col2
#>   <chr>  <chr>       <dbl>      <dbl>
#> 1 apple  aple            1          4
#> 2 banana banan           4          1
#> 3 banana bananb          4          1
#> 4 banana banat           4          1

^{由 reprex package (v2.0.1)}

创建于 2022-02-17

我想 group_by col1 并且当 count_col2 > count_col1 将 col1 的值转换为 NA，

当 count_col1 > count_col2 时，col2 的值将转换为 NA。

我希望我的数据看起来像这样

#> # A tibble: 4 × 4
#>   col1    col2        count_col1 count_col2
#>   <chr>   <chr>          <dbl>      <dbl>
#> 1   NA    aple             1          4
#> 2 banana   NA              4          1
#> 3 banana   NA              4          1
#> 4 banana   NA              4          1

我不确定这是否可以通过 mutate(case_when...) 到目前为止我失败了

df3 %>% 
  group_by(col1) %>% 
  mutate(case_when(count_col2 > count_col1 ~ col1==NA,
                   count_col1 > count_col2 ~ col2==NA ))

Answer 1

您可以使用 ifelse() 实现所需的输出，即

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

df3 <- tibble(col1 = c("apple",rep("banana",3)),
              col2 = c("aple", "banan","bananb","banat"), 
              count_col1 = c(1,4,4,4), 
              count_col2 = c(4,1,1,1))
df3
#> # A tibble: 4 × 4
#>   col1   col2   count_col1 count_col2
#>   <chr>  <chr>       <dbl>      <dbl>
#> 1 apple  aple            1          4
#> 2 banana banan           4          1
#> 3 banana bananb          4          1
#> 4 banana banat           4          1

df3 %>% 
  group_by(col1) %>% 
  mutate(col1 = ifelse(count_col2 > count_col1, NA, col1),
         col2 = ifelse(count_col1 > count_col2, NA, col2))
#> # A tibble: 4 × 4
#> # Groups:   col1 [2]
#>   col1   col2  count_col1 count_col2
#>   <chr>  <chr>      <dbl>      <dbl>
#> 1 <NA>   aple           1          4
#> 2 banana <NA>           4          1
#> 3 banana <NA>           4          1
#> 4 banana <NA>           4          1

^{由 reprex package (v2.0.1)}

于 2022-02-18 创建

或 case_when():

library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

df3 <- tibble(col1 = c("apple",rep("banana",3)),
              col2 = c("aple", "banan","bananb","banat"), 
              count_col1 = c(1,4,4,4), 
              count_col2 = c(4,1,1,1))
df3
#> # A tibble: 4 × 4
#>   col1   col2   count_col1 count_col2
#>   <chr>  <chr>       <dbl>      <dbl>
#> 1 apple  aple            1          4
#> 2 banana banan           4          1
#> 3 banana bananb          4          1
#> 4 banana banat           4          1

df3 %>% 
  group_by(col1) %>% 
  mutate(col1 = case_when(count_col2 > count_col1 ~ NA_character_,
                          TRUE ~ col1),
         col2 = case_when(count_col1 > count_col2 ~ NA_character_, 
                          TRUE ~ col2))
#> # A tibble: 4 × 4
#> # Groups:   col1 [2]
#>   col1   col2  count_col1 count_col2
#>   <chr>  <chr>      <dbl>      <dbl>
#> 1 <NA>   aple           1          4
#> 2 banana <NA>           4          1
#> 3 banana <NA>           4          1
#> 4 banana <NA>           4          1

^{由 reprex package (v2.0.1)}

于 2022-02-18 创建

这是否解决了您的问题？

Answer 2

我不确定您是否真的需要在此处 group_by，因为即使使用 group_by，count_col1 的每个值都会与 count_col2 的相应值进行比较。在“组”中没有发生任何事情。

这是一个基本的 R 选项 -

df3$col1[df3$count_col2 > df3$count_col1] <- NA
df3$col2[df3$count_col1 > df3$count_col2] <- NA
df3

#  col1   col2  count_col1 count_col2
#  <chr>  <chr>      <dbl>      <dbl>
#1 NA     aple           1          4
#2 banana NA             4          1
#3 banana NA             4          1
#4 banana NA             4          1

根据 R dplyr 中的条件将字符转换为 NA

Convert a character to NA based on a condition in R dplyr

r

dplyr

data.table

tidyr

tidyverse