brier_class: Brier score for classification models

In yardstick: Tidy Characterizations of Model Performance

Brier score for classification models

Description

Compute the Brier score for a classification model.

Usage

brier_class(data, ...)

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

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

Arguments

data	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.
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.
na_rm	A logical value indicating whether NA values should be stripped before the computation proceeds.
event_level	A single string. Either "first" or "second" to specify which level of truth to consider as the "event". This argument is only applicable when estimator = "binary". The default uses an internal helper that defaults to "first".
case_weights	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().
estimate	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.

Details

Brier score is a metric that should be minimized. The output ranges from 0 to 1, with 0 indicating perfect predictions.

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.

The formula used here is:

\text{Brier} = \frac{1}{2N} \sum_{i=1}^{N} \sum_{j=1}^{K} (y_{ij} - p_{ij})^2

where N is the number of observations, K is the number of classes, y_{ij} is 1 if observation i belongs to class j and 0 otherwise, and p_{ij} is the predicted probability of observation i for class j.

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.

Value

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).

Multiclass

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

Relevant Level

There is no common convention on which factor level should automatically be considered the "event" or "positive" result when computing binary classification metrics. In yardstick, the default is to use the first level. To alter this, change the argument event_level to "second" to consider the last level of the factor the level of interest. 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.

Author(s)

Max Kuhn

References

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

All probability metrics

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

Examples

# Two class
data("two_class_example")
brier_class(two_class_example, truth, Class1)

# Multiclass
library(dplyr)
data(hpc_cv)

# You can use the col1:colN tidyselect syntax
hpc_cv |>
  filter(Resample == "Fold01") |>
  brier_class(obs, VF:L)

# Groups are respected
hpc_cv |>
  group_by(Resample) |>
  brier_class(obs, VF:L)

yardstick documentation built on April 8, 2026, 1:06 a.m.