Identify optimal threshold

Retrieves the score threshold that maximizes F-score, which is a tradeoff between precision and recall.

Usage

wt_get_threshold(data)

Arguments

data

Tibble output from the wt_evaluate_classifier() function.

Value

A single numeric value

Examples

if (FALSE) { # \dontrun{
data <- wt_download_report(project_id = 1144, sensor_id = "ARU",
reports = c("main", "birdnet"), weather_cols = FALSE)

eval <- wt_evaluate_classifier(data, resolution = "recording",
remove_species = TRUE, thresholds = c(10, 99))

threshold_use <- wt_get_threshold(eval) |> print()
} # }