Calculate the symmetric mean absolute percentage error. This metric is in relative units.
smape(data, ...) # S3 method for data.frame smape(data, truth, estimate, na_rm = TRUE, ...) smape_vec(truth, estimate, na_rm = TRUE, ...)
data | A |
---|---|
... | Not currently used. |
truth | The column identifier for the true results
(that is |
estimate | The column identifier for the predicted
results (that is also |
na_rm | A |
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 smape_vec()
, a single numeric
value (or NA
).
This implementation of smape()
is the "usual definition" where the
denominator is divided by two.
Other numeric metrics:
ccc()
,
huber_loss_pseudo()
,
huber_loss()
,
iic()
,
mae()
,
mape()
,
mase()
,
mpe()
,
msd()
,
rmse()
,
rpd()
,
rpiq()
,
rsq_trad()
,
rsq()
Other accuracy metrics:
ccc()
,
huber_loss_pseudo()
,
huber_loss()
,
iic()
,
mae()
,
mape()
,
mase()
,
mpe()
,
msd()
,
rmse()
Max Kuhn, Riaz Hedayati
# Supply truth and predictions as bare column names smape(solubility_test, solubility, prediction)#> # A tibble: 1 x 3 #> .metric .estimator .estimate #> <chr> <chr> <dbl> #> 1 smape standard 36.7library(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) %>% smape(solubility, prediction) metric_results#> # A tibble: 10 x 4 #> resample .metric .estimator .estimate #> <chr> <chr> <chr> <dbl> #> 1 1 smape standard 33.5 #> 2 10 smape standard 28.9 #> 3 2 smape standard 41.6 #> 4 3 smape standard 28.4 #> 5 4 smape standard 37.0 #> 6 5 smape standard 31.8 #> 7 6 smape standard 48.3 #> 8 7 smape standard 27.8 #> 9 8 smape standard 35.6 #> 10 9 smape standard 30.2#> # A tibble: 1 x 1 #> avg_estimate #> <dbl> #> 1 34.3