Psuedo-Huber Loss

Calculate the Pseudo-Huber Loss, a smooth approximation of huber_loss(). Like huber_loss(), this is less sensitive to outliers than rmse().

huber_loss_pseudo(data, ...)

# S3 method for data.frame
huber_loss_pseudo(data, truth, estimate, delta = 1, na_rm = TRUE, ...)

huber_loss_pseudo_vec(truth, estimate, delta = 1, na_rm = TRUE, ...)

Arguments

data	A data.frame containing the truth and estimate columns.
...	Not currently used.
truth	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.
estimate	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.
delta	A single numeric value. Defines the boundary where the loss function transitions from quadratic to linear. Defaults to 1.
na_rm	A logical value indicating whether NA values should be stripped before the computation proceeds.

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

References

Huber, P. (1964). Robust Estimation of a Location Parameter. Annals of Statistics, 53 (1), 73-101.

Hartley, Richard (2004). Multiple View Geometry in Computer Vision. (Second Edition). Page 619.

See also

Other numeric metrics: ccc(), huber_loss(), iic(), mae(), mape(), mase(), rmse(), rpd(), rpiq(), rsq_trad(), rsq(), smape()

Other accuracy metrics: ccc(), huber_loss(), iic(), mae(), mape(), mase(), rmse(), smape()

Examples

# Supply truth and predictions as bare column names
huber_loss_pseudo(solubility_test, solubility, prediction)
#> # A tibble: 1 x 3
#>   .metric           .estimator .estimate
#>   <chr>             <chr>          <dbl>
#> 1 huber_loss_pseudo standard       0.199

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) %>%
  huber_loss_pseudo(solubility, prediction)

metric_results
#> # A tibble: 10 x 4
#>    resample .metric           .estimator .estimate
#>    <chr>    <chr>             <chr>          <dbl>
#>  1 1        huber_loss_pseudo standard       0.185
#>  2 10       huber_loss_pseudo standard       0.179
#>  3 2        huber_loss_pseudo standard       0.196
#>  4 3        huber_loss_pseudo standard       0.168
#>  5 4        huber_loss_pseudo standard       0.212
#>  6 5        huber_loss_pseudo standard       0.177
#>  7 6        huber_loss_pseudo standard       0.246
#>  8 7        huber_loss_pseudo standard       0.227
#>  9 8        huber_loss_pseudo standard       0.161
#> 10 9        huber_loss_pseudo standard       0.188

# Resampled mean estimate
metric_results %>%
  summarise(avg_estimate = mean(.estimate))
#> # A tibble: 1 x 1
#>   avg_estimate
#>          <dbl>
#> 1        0.194