R yardstick roc_aunu 使用均勻類別分布，每個類別相對於其他類別的 ROC 曲線下麵積

roc_aunu(data, ...)

# S3 method for data.frame
roc_aunu(data, truth, ..., na_rm = TRUE, case_weights = NULL, options = list())

roc_aunu_vec(
  truth,
  estimate,
  na_rm = TRUE,
  case_weights = NULL,
  options = list(),
  ...
)

roc_aunp() 用於使用先驗類別分布計算每個類別相對於其他類別的 ROC 曲線下麵積。

其他類概率指標：average_precision() , brier_class() , classification_cost() , gain_capture() , mn_log_loss() , pr_auc() , roc_auc() , roc_aunp()

# Multiclass example

# `obs` is a 4 level factor. The first level is `"VF"`, which is the
# "event of interest" by default in yardstick. See the Relevant Level
# section above.
data(hpc_cv)

# You can use the col1:colN tidyselect syntax
library(dplyr)
hpc_cv %>%
  filter(Resample == "Fold01") %>%
  roc_aunu(obs, VF:L)
#> # A tibble: 1 × 3
#>   .metric  .estimator .estimate
#>   <chr>    <chr>          <dbl>
#> 1 roc_aunu macro          0.871

# Change the first level of `obs` from `"VF"` to `"M"` to alter the
# event of interest. The class probability columns should be supplied
# in the same order as the levels.
hpc_cv %>%
  filter(Resample == "Fold01") %>%
  mutate(obs = relevel(obs, "M")) %>%
  roc_aunu(obs, M, VF:L)
#> # A tibble: 1 × 3
#>   .metric  .estimator .estimate
#>   <chr>    <chr>          <dbl>
#> 1 roc_aunu macro          0.871

# Groups are respected
hpc_cv %>%
  group_by(Resample) %>%
  roc_aunu(obs, VF:L)
#> # A tibble: 10 × 4
#>    Resample .metric  .estimator .estimate
#>    <chr>    <chr>    <chr>          <dbl>
#>  1 Fold01   roc_aunu macro          0.871
#>  2 Fold02   roc_aunu macro          0.863
#>  3 Fold03   roc_aunu macro          0.898
#>  4 Fold04   roc_aunu macro          0.874
#>  5 Fold05   roc_aunu macro          0.865
#>  6 Fold06   roc_aunu macro          0.877
#>  7 Fold07   roc_aunu macro          0.865
#>  8 Fold08   roc_aunu macro          0.873
#>  9 Fold09   roc_aunu macro          0.855
#> 10 Fold10   roc_aunu macro          0.865

# Vector version
# Supply a matrix of class probabilities
fold1 <- hpc_cv %>%
  filter(Resample == "Fold01")

roc_aunu_vec(
  truth = fold1$obs,
  matrix(
    c(fold1$VF, fold1$F, fold1$M, fold1$L),
    ncol = 4
  )
)
#> [1] 0.8714461