次の方法で共有


DataDriftMonitoringSignal コンストラクター

定義

DataDriftMonitoringSignal の新しいインスタンスを初期化します。

public DataDriftMonitoringSignal (System.Collections.Generic.IEnumerable<Azure.ResourceManager.MachineLearning.Models.DataDriftMetricThresholdBase> metricThresholds, Azure.ResourceManager.MachineLearning.Models.MonitoringInputDataBase productionData, Azure.ResourceManager.MachineLearning.Models.MonitoringInputDataBase referenceData);
new Azure.ResourceManager.MachineLearning.Models.DataDriftMonitoringSignal : seq<Azure.ResourceManager.MachineLearning.Models.DataDriftMetricThresholdBase> * Azure.ResourceManager.MachineLearning.Models.MonitoringInputDataBase * Azure.ResourceManager.MachineLearning.Models.MonitoringInputDataBase -> Azure.ResourceManager.MachineLearning.Models.DataDriftMonitoringSignal
Public Sub New (metricThresholds As IEnumerable(Of DataDriftMetricThresholdBase), productionData As MonitoringInputDataBase, referenceData As MonitoringInputDataBase)

パラメーター

metricThresholds
IEnumerable<DataDriftMetricThresholdBase>

[必須]計算するメトリックと、関連するしきい値の一覧。 基底クラスであることに注意 DataDriftMetricThresholdBase してください。 このシナリオでは、基底クラスの派生クラスをここで割り当てる必要があるか、このプロパティを使用可能な派生クラスのいずれかにキャストする必要があります。 使用可能な派生クラスには、 と がNumericalDataDriftMetricThreshold含まれますCategoricalDataDriftMetricThreshold

productionData
MonitoringInputDataBase

[必須]誤差が計算されるデータ。 基底クラスであることに注意 MonitoringInputDataBase してください。 このシナリオでは、基底クラスの派生クラスをここで割り当てる必要があるか、このプロパティを使用可能な派生クラスのいずれかにキャストする必要があります。 使用可能な派生クラスには、、StaticInputDataおよび TrailingInputDataが含まれますFixedInputData

referenceData
MonitoringInputDataBase

[必須]ドリフトを計算するデータ。 基底クラスであることに注意 MonitoringInputDataBase してください。 このシナリオでは、基底クラスの派生クラスをここで割り当てる必要があるか、このプロパティを使用可能な派生クラスのいずれかにキャストする必要があります。 使用可能な派生クラスには、、StaticInputDataおよび TrailingInputDataが含まれますFixedInputData

例外

metricThresholdsproductionData、または referenceData が null です。

適用対象