Class SmartDetectionCondition
java.lang.Object
com.azure.ai.metricsadvisor.administration.models.SmartDetectionCondition
Type that describes smart-detection parameters. In smart-detection mode, metrics advisor
uses multiple ML based algorithms to compute severity value of data points, detector detect
anomalies by checking whether those values falls within or outside of the range derived from
sensitivity parameter.
-
Constructor Summary
ConstructorDescriptionSmartDetectionCondition
(double sensitivity, AnomalyDetectorDirection detectorDirection, SuppressCondition suppressCondition) Create an instance of SmartDetectionCondition describing how to identify anomalies using smart-detection mode. -
Method Summary
Modifier and TypeMethodDescriptionGets the direction that detector should use when comparing data point value against range derived from sensitivity .Gets the sensitivity value.Gets the suppress condition.setAnomalyDetectorDirection
(AnomalyDetectorDirection detectorDirection) Sets the direction that detector should use when comparing data point value against range derived from the sensitivity.setSensitivity
(double sensitivity) Sets the sensitivity value, it should be in the range (0, 100].setSuppressCondition
(SuppressCondition suppressCondition) Sets the suppress condition.
-
Constructor Details
-
SmartDetectionCondition
public SmartDetectionCondition(double sensitivity, AnomalyDetectorDirection detectorDirection, SuppressCondition suppressCondition) Create an instance of SmartDetectionCondition describing how to identify anomalies using smart-detection mode.- Parameters:
sensitivity
- value that adjust the tolerance of anomalies, visually higher the value narrower the band (lower and upper bounds) around the time series.detectorDirection
- a valueAnomalyDetectorDirection.BOTH
indicates that any data point with severity value not within the range derived fromsensitivity
should be considered as an anomaly. A valueAnomalyDetectorDirection.UP
means a data point severity value above the upper bound of the range is considered as an anomaly, a valueAnomalyDetectorDirection.DOWN
means a data point severity value below lower bound of the range is considered as an anomaly.suppressCondition
- the condition to aggregate the anomaly detection reporting, suppressing the reporting of individual anomalies helps to avoid noises, especially if the metrics have fine granularity.
-
-
Method Details
-
getSensitivity
Gets the sensitivity value.the sensitivity value adjust the tolerance of anomalies, visually higher the value narrower the band (lower and upper bounds) around the time series.
- Returns:
- the sensitivity value.
-
getAnomalyDetectorDirection
Gets the direction that detector should use when comparing data point value against range derived from sensitivity .- Returns:
- the detector direction.
-
getSuppressCondition
Gets the suppress condition.- Returns:
- the suppress condition value.
-
setSensitivity
Sets the sensitivity value, it should be in the range (0, 100].- Parameters:
sensitivity
- the sensitivity value to set.- Returns:
- the SmartDetectionCondition object itself.
-
setAnomalyDetectorDirection
public SmartDetectionCondition setAnomalyDetectorDirection(AnomalyDetectorDirection detectorDirection) Sets the direction that detector should use when comparing data point value against range derived from the sensitivity.- Parameters:
detectorDirection
- the detector direction- Returns:
- the HardThresholdCondition object itself.
-
setSuppressCondition
Sets the suppress condition.- Parameters:
suppressCondition
- the suppress condition- Returns:
- the HardThresholdCondition object itself.
-