Threshold value location for classification

[erikerhardt]

rfsrc team,

I'm looking for the threshold value used for the confusion matrix printed in the standard output from the rfsrc object. Where is this threshold value located?

For imbalanced(), I can get thresholds with these functions, but their thresholds result in different confusion matrices.

    ## (1) default threshold (2) directly optimized gmean threshold
    threshold_default <-
      randomForestSRC::get.imbalanced.performance(
        o_class_full
      )["threshold"]

    threshold_gmean   <-
      randomForestSRC::get.imbalanced.optimize(
        o_class_full
      , measure = c("gmean", "F1", "F1mod", "F1modgmean")[1]
      , plot.it = FALSE
      )["threshold"] |>
      as.numeric()

Confusion matrix:

          predicted
  observed   No Yes class.error
       No  8526  18      0.0021
       Yes  172 367      0.3191

Thank you!