Tidy model metrics
Usage
metrics(
x,
model = NULL,
scope = NULL,
subject = NULL,
n = NULL,
overall = FALSE,
sort_by = NULL,
decreasing = FALSE,
by = c("person", "fold"),
...
)Arguments
- x
An idiographic fit.
- model, scope, subject
Optional filters.
- n
Optional number of rows to return.
- overall
Logical. If
TRUE, return only overall rows.- sort_by
Optional metric/column to sort by.
- decreasing
Sort order when
sort_byis supplied.- by
Summarise by person (default) or separately within each rolling fold.
- ...
Ignored.
Value
A data.frame with one row per scope, model, estimator and person
(columns scope, model, estimator, subject, subgroup, n), plus
one .overall row per model that pools all held-out occasions. Regression
fits report rmse, mae, bias and r_squared; classification fits
report accuracy, brier and log_loss. With by = "fold", rolling fits add a
fold column and report each fold separately.
Examples
fit <- fit_lm(srl, y = "effort", x = "efficacy", id = "name", time = "day")
metrics(fit, overall = TRUE)
#> scope model estimator subject subgroup n rmse mae bias
#> 1 pooled lm native .overall .all 1150 22.15170 18.16606 0.5861259
#> 2 individual lm native .overall .all 1150 19.22193 14.95209 0.4451146
#> r_squared
#> 1 0.3200665
#> 2 0.4880281
# The five learners the person-specific model predicts worst.
metrics(fit, scope = "individual", sort_by = "rmse", decreasing = TRUE,
n = 5)
#> scope model estimator subject subgroup n rmse mae bias
#> 1 individual lm native Erik .none 32 29.73167 26.93958 0.8066997
#> 2 individual lm native Frank .none 32 28.54220 25.28387 0.8046534
#> 3 individual lm native Eve .none 32 27.67821 23.65575 -5.5625396
#> 4 individual lm native Yuki .none 32 24.56427 21.67633 0.6466954
#> 5 individual lm native Layla .none 30 24.49670 17.99006 3.7420751
#> r_squared
#> 1 0.35044875
#> 2 0.01234187
#> 3 0.34040623
#> 4 0.09619659
#> 5 0.07071314