`R/prob-roc_aunp.R`

`roc_aunp.Rd`

`roc_aunp()`

is a multiclass metric that computes the area under the ROC
curve of each class against the rest, using the a priori class distribution.
This is equivalent to `roc_auc(estimator = "macro_weighted")`

.

roc_aunp(data, ...) # S3 method for data.frame roc_aunp(data, truth, ..., options = list(), na_rm = TRUE) roc_aunp_vec(truth, estimate, options = list(), na_rm = TRUE, ...)

data | A |
---|---|

... | A set of unquoted column names or one or more |

truth | The column identifier for the true class results
(that is a |

options | A |

na_rm | A |

estimate | A matrix with as many
columns as factor levels of |

A `tibble`

with columns `.metric`

, `.estimator`

,
and `.estimate`

and 1 row of values.

For grouped data frames, the number of rows returned will be the same as the number of groups.

For `roc_aunp_vec()`

, a single `numeric`

value (or `NA`

).

Like the other ROC AUC metrics, `roc_aunp()`

defaults to allowing
`pROC::roc()`

control the direction of the computation, but allows you to
control this by passing `options = list(direction = "<")`

or any other
allowed direction value. pROC advises setting the `direction`

when doing
resampling so that the AUC values are not biased upwards.

Generally, an ROC AUC value is between `0.5`

and `1`

, with `1`

being a
perfect prediction model. If your value is between `0`

and `0.5`

, then
this implies that you have meaningful information in your model, but it
is being applied incorrectly because doing the opposite of what the model
predicts would result in an AUC `>0.5`

.

There is no common convention on which factor level should
automatically be considered the "event" or "positive" result.
In `yardstick`

, the default is to use the *first* level. To
change this, a global option called `yardstick.event_first`

is
set to `TRUE`

when the package is loaded. This can be changed
to `FALSE`

if the *last* level of the factor is considered the
level of interest by running: `options(yardstick.event_first = FALSE)`

.
For multiclass extensions involving one-vs-all
comparisons (such as macro averaging), this option is ignored and
the "one" level is always the relevant result.

This multiclass method for computing the area under the ROC curve uses the
a priori class distribution and is equivalent to
`roc_auc(estimator = "macro_weighted")`

.

Ferri, C., Hernández-Orallo, J., & Modroiu, R. (2009). "An experimental
comparison of performance measures for classification". *Pattern Recognition
Letters*. 30 (1), pp 27-38.

`roc_aunu()`

for computing the area under the ROC curve of each class against
the rest, using the uniform class distribution.

Other class probability metrics:
`average_precision()`

,
`gain_capture()`

,
`mn_log_loss()`

,
`pr_auc()`

,
`roc_auc()`

,
`roc_aunu()`

# 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_aunp(obs, VF:L)#> # A tibble: 1 x 3 #> .metric .estimator .estimate #> <chr> <chr> <dbl> #> 1 roc_aunp macro 0.880# 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_aunp(obs, M, VF:L)#> # A tibble: 1 x 3 #> .metric .estimator .estimate #> <chr> <chr> <dbl> #> 1 roc_aunp macro 0.880#> # A tibble: 10 x 4 #> Resample .metric .estimator .estimate #> <chr> <chr> <chr> <dbl> #> 1 Fold01 roc_aunp macro 0.880 #> 2 Fold02 roc_aunp macro 0.873 #> 3 Fold03 roc_aunp macro 0.906 #> 4 Fold04 roc_aunp macro 0.867 #> 5 Fold05 roc_aunp macro 0.866 #> 6 Fold06 roc_aunp macro 0.865 #> 7 Fold07 roc_aunp macro 0.868 #> 8 Fold08 roc_aunp macro 0.865 #> 9 Fold09 roc_aunp macro 0.841 #> 10 Fold10 roc_aunp macro 0.869# Vector version # Supply a matrix of class probabilities fold1 <- hpc_cv %>% filter(Resample == "Fold01") roc_aunp_vec( truth = fold1$obs, matrix( c(fold1$VF, fold1$F, fold1$M, fold1$L), ncol = 4 ) )#> [1] 0.8795121# --------------------------------------------------------------------------- # Options for `pROC::roc()` # Pass options via a named list and not through `...`! roc_aunp( hpc_cv, obs, VF:L, options = list(smooth = TRUE) )#> # A tibble: 1 x 3 #> .metric .estimator .estimate #> <chr> <chr> <dbl> #> 1 roc_aunp macro 0.868