This function allows to use a custom thresholding method to maximize the Maximum F1 Score (Precision with Sensitivity harmonic mean).
metrics.f1.max(preds, labels, threshold = FALSE)
| preds | Type: numeric. The predictions. |
|---|---|
| labels | Type: numeric. The labels (0, 1). |
| threshold | Type: logical. Whether to return the threshold. Defaults to |
A one or two element vector containing the maximum Maximum F1 Score (Precision with Sensitivity harmonic mean) for binary data, and the threshold used if requested. Returns -1 in case of errors.