Skip to content

Summary Statistics for Confusion Matrices

Source: R/conf_mat.R
summary.conf_mat.Rd

Various statistical summaries of confusion matrices are produced and returned in a tibble. These include those shown in the help pages for sens(), recall(), and accuracy(), among others.

Usage

# S3 method for conf_mat
summary(
  object,
  prevalence = NULL,
  beta = 1,
  estimator = NULL,
  event_level = yardstick_event_level(),
  ...
)

Arguments

object

An object of class conf_mat().

prevalence

A number in (0, 1) for the prevalence (i.e. prior) of the event. If left to the default, the data are used to derive this value.

beta

A numeric value used to weight precision and recall for f_meas().

estimator

One of: "binary", "macro", "macro_weighted", or "micro" to specify the type of averaging to be done. "binary" is only relevant for the two class case. The other three are general methods for calculating multiclass metrics. The default will automatically choose "binary" or "macro" based on estimate.

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".

...

Not currently used.

Value

A tibble containing various classification metrics.

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.

See also

conf_mat()

Examples

data("two_class_example")

cmat <- conf_mat(two_class_example, truth = "truth", estimate = "predicted")
summary(cmat)
#> # A tibble: 13 × 3
#>    .metric              .estimator .estimate
#>    <chr>                <chr>          <dbl>
#>  1 accuracy             binary         0.838
#>  2 kap                  binary         0.675
#>  3 sens                 binary         0.880
#>  4 spec                 binary         0.793
#>  5 ppv                  binary         0.819
#>  6 npv                  binary         0.861
#>  7 mcc                  binary         0.677
#>  8 j_index              binary         0.673
#>  9 bal_accuracy         binary         0.837
#> 10 detection_prevalence binary         0.554
#> 11 precision            binary         0.819
#> 12 recall               binary         0.880
#> 13 f_meas               binary         0.849
summary(cmat, prevalence = 0.70)
#> # A tibble: 13 × 3
#>    .metric              .estimator .estimate
#>    <chr>                <chr>          <dbl>
#>  1 accuracy             binary         0.838
#>  2 kap                  binary         0.675
#>  3 sens                 binary         0.880
#>  4 spec                 binary         0.793
#>  5 ppv                  binary         0.909
#>  6 npv                  binary         0.739
#>  7 mcc                  binary         0.677
#>  8 j_index              binary         0.673
#>  9 bal_accuracy         binary         0.837
#> 10 detection_prevalence binary         0.554
#> 11 precision            binary         0.819
#> 12 recall               binary         0.880
#> 13 f_meas               binary         0.849

library(dplyr)
library(tidyr)
data("hpc_cv")

# Compute statistics per resample then summarize
all_metrics <- hpc_cv %>%
  group_by(Resample) %>%
  conf_mat(obs, pred) %>%
  mutate(summary_tbl = lapply(conf_mat, summary)) %>%
  unnest(summary_tbl)

all_metrics %>%
  group_by(.metric) %>%
  summarise(
    mean = mean(.estimate, na.rm = TRUE),
    sd = sd(.estimate, na.rm = TRUE)
  )
#> # A tibble: 13 × 3
#>    .metric               mean       sd
#>    <chr>                <dbl>    <dbl>
#>  1 accuracy             0.709 2.47e- 2
#>  2 bal_accuracy         0.720 1.92e- 2
#>  3 detection_prevalence 0.25  9.25e-18
#>  4 f_meas               0.569 3.46e- 2
#>  5 j_index              0.439 3.85e- 2
#>  6 kap                  0.508 4.10e- 2
#>  7 mcc                  0.515 4.16e- 2
#>  8 npv                  0.896 1.11e- 2
#>  9 ppv                  0.633 3.87e- 2
#> 10 precision            0.633 3.87e- 2
#> 11 recall               0.560 3.09e- 2
#> 12 sens                 0.560 3.09e- 2
#> 13 spec                 0.879 9.67e- 3