在dplyr的mutate中使用switch语句

时间:2017-10-18 15:38:07

标签: r switch-statement dplyr mutate

我想在dplyr的mutate中使用switch语句。我有一个简单的函数,它执行一些操作并通过switch指定替代值,例如:

convert_am <- function(x) {
    x <- as.character(x)
    switch(x,
           "0" = FALSE,
           "1" = TRUE,
           NA)
}

当应用于标量时,这可以正常工作:

>> convert_am(1)
[1] TRUE
>> convert_am(2)
[1] NA
>> convert_am(0)
[1] FALSE

我希望通过mutate调用来获得相同的结果:

mtcars %>% mutate(am = convert_am(am))

这失败了:

  

mutate_impl(.data, dots)出错:评估错误:EXPR必须   是长度为1的向量。

据我所知,这是因为传递给switch ar的值不是单一的,例如:

  {p> convert_am(c(1,2,2)) switch(x, 0 = FALSE, 1 = TRUE, NA)出错:EXPR必须是长度为1的向量

矢量

尝试矢量化也会产生预期的结果:

convert_am <- function(x) {
    x <- as.character(x)

    fun_switch <- function(x) {
        switch(x,
               "0" = FALSE,
               "1" = TRUE,
               NA)
    }

    vf <- Vectorize(fun_switch, "x")
}

>> mtcars %>% mutate(am = convert_am(am))
Error in mutate_impl(.data, dots) : 
  Column `am` is of unsupported type function

注释

  • 我知道dplyr中的case_when并且我对使用它不感兴趣,我只对使switch在mutate中工作感兴趣
  • 理想的解决方案可以进一步扩展,以便将变量作为 mutate_at
  • 传递给.

2 个答案:

答案 0 :(得分:5)

switch未进行矢量化,因此为了提高效率,您需要使用ifelsecase_when - 但由于您的问题特别针对switch,因此您可以实现所需的效果矢量化,例如

convert_am <- Vectorize(function(x) {
    x <- as.character(x)
    switch(x,
       "0" = FALSE,
       "1" = TRUE,
       NA)
})

convert_am <- function(x) {
    x <- as.character(x)
    sapply(x, function(xx) switch(xx,
       "0" = FALSE,
       "1" = TRUE,
       NA))
}

它们效率低下,因为它们涉及到引擎盖下的循环。

答案 1 :(得分:1)

This is simple enough to handle with ifelse directly:
        
            Test <- tibble::tibble(
              am = c(-1:5,NA,1, 0)
            ) 
        
            Test %>%
              mutate(
                newam = ifelse(am == 1, TRUE,
                       ifelse(am == 0, FALSE, NA))
              )
        
        
With more categories, use a named vector:
            Test %>%
              mutate(
                newam = ifelse(is.na(am) | !am %in% c(1,3,5), NA,
                               c("1" = "in1", "3" = "in3", "5" = "in5")[as.character(am)])
              )
        
In fact if the value is not in the named list it will default to an NA
I think this will be pretty efficient
            Test %>%
              mutate(
                newam = c("1" = "in1", "3" = "in3", "5" = "in5")[as.character(am)]
              )