1004 B
1004 B
MlEvaluateDataFrameDataframeOutlierDetectionSummary [interface-MlEvaluateDataFrameDataframeOutlierDetectionSummary]
| Name | Type | Description |
|---|---|---|
auc_roc |
MlEvaluateDataFrameDataframeEvaluationSummaryAucRoc | The AUC ROC (area under the curve of the receiver operating characteristic) score and optionally the curve. |
confusion_matrix |
Record<string, MlEvaluateDataFrameConfusionMatrixThreshold> | Set the different thresholds of the outlier score at where the metrics ( tp - true positive, fp - false positive, tn - true negative, fn - false negative) are calculated. |
precision |
Record<string, double> | Set the different thresholds of the outlier score at where the metric is calculated. |
recall |
Record<string, double> | Set the different thresholds of the outlier score at where the metric is calculated. |