Convenience helpers to expose overall metrics, cross-validation summaries, and per-class statistics in both tabular and text-friendly formats.
Value
`rf_metric_table()` returns a list of tables: `overall`, `cv_overall`, `cv_overall_summary`, `class`, and `class_raw`. `rf_metric_text()` returns a character vector of summary lines.
Examples
fit <- list(
metrics = list(
accuracy = 0.75,
f1_macro = 0.73,
cv_overall = data.frame(
accuracy = c(0.7, 0.8),
f1_macro = c(0.68, 0.78)
)
),
test_truth = c("Bcell", "Bcell", "Tcell", "Tcell"),
test_pred = c("Bcell", "Tcell", "Tcell", "Tcell")
)
rf_metric_table(fit)
#> $overall
#> metric value
#> 1 accuracy 0.75
#> 2 f1_macro 0.73
#>
#> $cv_overall
#> accuracy f1_macro
#> 1 0.7 0.68
#> 2 0.8 0.78
#>
#> $cv_overall_summary
#> accuracy_mean accuracy_sd f1_macro_mean f1_macro_sd
#> 1 0.75 0.07071068 0.73 0.07071068
#>
#> $class
#> class tp fp fn support precision recall f1
#> Bcell Bcell 1 0 1 2 1.0000000 0.5 0.6666667
#> Tcell Tcell 2 1 0 2 0.6666667 1.0 0.8000000
#>
#> $class_raw
#> class tp fp fn support precision recall f1
#> Bcell Bcell 1 0 1 2 1.0000000 0.5 0.6666667
#> Tcell Tcell 2 1 0 2 0.6666667 1.0 0.8000000
#>