Computes AUC-ROC, AUC-PR, and Top-K Recall metrics for evaluating
anomaly detection performance against ground truth.
Usage
calculate_benchmark_metrics(scores, ground_truth, contamination = 0.05)
Arguments
- scores
Numeric vector of anomaly scores
- ground_truth
Binary vector (0/1) of true anomaly labels
- contamination
Expected proportion of anomalies
Value
List of benchmarking metrics
