Calculate the mean percentage error. This metric is in relative
units. It can be used as a measure of the estimate's bias.
Note that if any truth values are 0, a value of:
-Inf (estimate > 0), Inf (estimate < 0), or NaN (estimate == 0)
is returned for mpe().
mpe(data, ...)
# S3 method for class 'data.frame'
mpe(data, truth, estimate, na_rm = TRUE, case_weights = NULL, ...)
mpe_vec(truth, estimate, na_rm = TRUE, case_weights = NULL, ...)A data.frame containing the columns specified by the truth
and estimate arguments.
Not currently used.
The column identifier for the true results
(that is numeric). 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 numeric vector.
The column identifier for the predicted
results (that is also numeric). As with truth this can be
specified different ways but the primary method is to use an
unquoted variable name. For _vec() functions, a numeric 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().
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 mpe_vec(), a single numeric value (or NA).
Other numeric metrics:
ccc(),
huber_loss(),
huber_loss_pseudo(),
iic(),
mae(),
mape(),
mase(),
msd(),
poisson_log_loss(),
rmse(),
rpd(),
rpiq(),
rsq(),
rsq_trad(),
smape()
Other accuracy metrics:
ccc(),
huber_loss(),
huber_loss_pseudo(),
iic(),
mae(),
mape(),
mase(),
msd(),
poisson_log_loss(),
rmse(),
smape()
# `solubility_test$solubility` has zero values with corresponding
# `$prediction` values that are negative. By definition, this causes `Inf`
# to be returned from `mpe()`.
solubility_test[solubility_test$solubility == 0, ]
#> solubility prediction
#> 17 0 -0.1532030
#> 220 0 -0.3876578
mpe(solubility_test, solubility, prediction)
#> # A tibble: 1 × 3
#> .metric .estimator .estimate
#> <chr> <chr> <dbl>
#> 1 mpe standard Inf
# We'll remove the zero values for demonstration
solubility_test <- solubility_test[solubility_test$solubility != 0, ]
# Supply truth and predictions as bare column names
mpe(solubility_test, solubility, prediction)
#> # A tibble: 1 × 3
#> .metric .estimator .estimate
#> <chr> <chr> <dbl>
#> 1 mpe standard 16.1
library(dplyr)
set.seed(1234)
size <- 100
times <- 10
# create 10 resamples
solubility_resampled <- bind_rows(
replicate(
n = times,
expr = sample_n(solubility_test, size, replace = TRUE),
simplify = FALSE
),
.id = "resample"
)
# Compute the metric by group
metric_results <- solubility_resampled %>%
group_by(resample) %>%
mpe(solubility, prediction)
metric_results
#> # A tibble: 10 × 4
#> resample .metric .estimator .estimate
#> <chr> <chr> <chr> <dbl>
#> 1 1 mpe standard -56.2
#> 2 10 mpe standard 50.4
#> 3 2 mpe standard -27.9
#> 4 3 mpe standard 0.470
#> 5 4 mpe standard -0.836
#> 6 5 mpe standard -35.3
#> 7 6 mpe standard 7.51
#> 8 7 mpe standard -34.5
#> 9 8 mpe standard 7.87
#> 10 9 mpe standard 14.7
# Resampled mean estimate
metric_results %>%
summarise(avg_estimate = mean(.estimate))
#> # A tibble: 1 × 1
#> avg_estimate
#> <dbl>
#> 1 -7.38