R recipes step_ordinalscore 将序数因子转换为数值分数

step_ordinalscore(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  columns = NULL,
  convert = as.numeric,
  skip = FALSE,
  id = rand_id("ordinalscore")
)

其他虚拟变量和编码步骤：step_bin2factor() , step_count() , step_date() , step_dummy_extract() , step_dummy_multi_choice() , step_dummy() , step_factor2string() , step_holiday() , step_indicate_na() , step_integer() , step_novel() , step_num2factor() , step_other() , step_regex() , step_relevel() , step_string2factor() , step_time() , step_unknown() , step_unorder()

fail_lvls <- c("meh", "annoying", "really_bad")

ord_data <-
  data.frame(
    item = c("paperclip", "twitter", "airbag"),
    fail_severity = factor(fail_lvls,
      levels = fail_lvls,
      ordered = TRUE
    )
  )

model.matrix(~fail_severity, data = ord_data)
#>   (Intercept) fail_severity.L fail_severity.Q
#> 1           1   -7.071068e-01       0.4082483
#> 2           1   -9.073800e-17      -0.8164966
#> 3           1    7.071068e-01       0.4082483
#> attr(,"assign")
#> [1] 0 1 1
#> attr(,"contrasts")
#> attr(,"contrasts")$fail_severity
#> [1] "contr.poly"
#> 

linear_values <- recipe(~ item + fail_severity, data = ord_data) %>%
  step_dummy(item) %>%
  step_ordinalscore(fail_severity)

linear_values <- prep(linear_values, training = ord_data)

bake(linear_values, new_data = NULL, everything())
#> # A tibble: 3 × 3
#>   fail_severity item_paperclip item_twitter
#>           <int>          <dbl>        <dbl>
#> 1             1              1            0
#> 2             2              0            1
#> 3             3              0            0

custom <- function(x) {
  new_values <- c(1, 3, 7)
  new_values[as.numeric(x)]
}

nonlin_scores <- recipe(~ item + fail_severity, data = ord_data) %>%
  step_dummy(item) %>%
  step_ordinalscore(fail_severity, convert = custom)

tidy(nonlin_scores, number = 2)
#> # A tibble: 1 × 2
#>   terms         id                
#>   <chr>         <chr>             
#> 1 fail_severity ordinalscore_eIAmG

nonlin_scores <- prep(nonlin_scores, training = ord_data)

bake(nonlin_scores, new_data = NULL, everything())
#> # A tibble: 3 × 3
#>   fail_severity item_paperclip item_twitter
#>           <int>          <dbl>        <dbl>
#> 1             1              1            0
#> 2             3              0            1
#> 3             7              0            0

tidy(nonlin_scores, number = 2)
#> # A tibble: 1 × 2
#>   terms         id                
#>   <chr>         <chr>             
#> 1 fail_severity ordinalscore_eIAmG