Performance metrics¶
The aeon.performance_metrics
module contains metrics for evaluating and tuning time series models.
Metrics for assessing model performance.
Segmentation¶
|
Error counting the difference in the number of change points. |
|
Compute the Hausdorff distance between two sets of change points. |
|
Prediction ratio is the ratio of number of predicted to true change points. |
Anomaly Detection¶
|
Compute the range-based precision metric. |
|
Compute the range-based recall metric. |
|
Compute the F-score using the range-based recall and precision metrics. |
|
Compute the ROC AUC score. |
|
Compute the precision-recall AUC score. |
|
Compute the AUC-score of the range-based precision-recall curve. |
|
Compute the F-score at k based on single points. |
|
Compute the range-based F-score at k based on anomaly ranges. |
|
Compute the range-based PR and ROC AUC. |
|
Compute the range-based area under the ROC curve. |
|
Compute the area under the range-based PR curve. |
|
Compute the range-based PR VUS score. |
|
Compute the range-based ROC VUS score. |
AD Thresholding¶
|
Calculate a threshold based on a percentile of the anomaly scores. |
|
Calculate a threshold based on the standard deviation of the anomaly scores. |
|
Calculate a threshold such that at least k anomalous points are found. |
|
Calculate a threshold such that at least k anomalies are found. |