Skip to content

Compute the Brier score for a classification model.


brier_class(data, ...)

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

brier_class_vec(truth, estimate, na_rm = TRUE, case_weights = NULL, ...)



A data.frame containing the columns specified by truth and ....


A set of unquoted column names or one or more dplyr selector functions to choose which variables contain the class probabilities. If truth is binary, only 1 column should be selected, and it should correspond to the value of event_level. Otherwise, there should be as many columns as factor levels of truth and the ordering of the columns should be the same as the factor levels of truth.


The column identifier for the true class results (that is a factor). This should be an unquoted column name although this argument is passed by expression and supports quasiquotation (you can unquote column names). For _vec() functions, a factor vector.


A logical value indicating whether NA values should be stripped before the computation proceeds.


The optional column identifier for case weights. This should be an unquoted column name that evaluates to a numeric column in data. For _vec() functions, a numeric vector, hardhat::importance_weights(), or hardhat::frequency_weights().


If truth is binary, a numeric vector of class probabilities corresponding to the "relevant" class. Otherwise, a matrix with as many columns as factor levels of truth. It is assumed that these are in the same order as the levels of truth.


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 brier_class_vec(), a single numeric value (or NA).


The Brier score is analogous to the mean squared error in regression models. The difference between a binary indicator for a class and its corresponding class probability are squared and averaged.

This function uses the convention in Kruppa et al (2014) and divides the result by two.

Smaller values of the score are associated with better model performance.


Brier scores can be computed in the same way for any number of classes. Because of this, no averaging types are supported.


Kruppa, J., Liu, Y., Diener, H.-C., Holste, T., Weimar, C., Koonig, I. R., and Ziegler, A. (2014) Probability estimation with machine learning methods for dichotomous and multicategory outcome: Applications. Biometrical Journal, 56 (4): 564-583.

See also

Other class probability metrics: average_precision(), classification_cost(), gain_capture(), mn_log_loss(), pr_auc(), roc_auc(), roc_aunp(), roc_aunu()


Max Kuhn


# Two class
brier_class(two_class_example, truth, Class1)
#> # A tibble: 1 × 3
#>   .metric     .estimator .estimate
#>   <chr>       <chr>          <dbl>
#> 1 brier_class binary         0.106

# Multiclass

# You can use the col1:colN tidyselect syntax
hpc_cv %>%
  filter(Resample == "Fold01") %>%
  brier_class(obs, VF:L)
#> # A tibble: 1 × 3
#>   .metric     .estimator .estimate
#>   <chr>       <chr>          <dbl>
#> 1 brier_class multiclass     0.202

# Groups are respected
hpc_cv %>%
  group_by(Resample) %>%
  brier_class(obs, VF:L)
#> # A tibble: 10 × 4
#>    Resample .metric     .estimator .estimate
#>    <chr>    <chr>       <chr>          <dbl>
#>  1 Fold01   brier_class multiclass     0.202
#>  2 Fold02   brier_class multiclass     0.215
#>  3 Fold03   brier_class multiclass     0.177
#>  4 Fold04   brier_class multiclass     0.204
#>  5 Fold05   brier_class multiclass     0.213
#>  6 Fold06   brier_class multiclass     0.214
#>  7 Fold07   brier_class multiclass     0.221
#>  8 Fold08   brier_class multiclass     0.209
#>  9 Fold09   brier_class multiclass     0.235
#> 10 Fold10   brier_class multiclass     0.218