Metodo GetAttributes
Gets a collection of attributes based on their feature selection status.
Spazio dei nomi: Microsoft.AnalysisServices.AdomdClient
Assembly: Microsoft.AnalysisServices.AdomdClient (in Microsoft.AnalysisServices.AdomdClient.dll)
Sintassi
'Dichiarazione
Public Function GetAttributes ( _
filter As MiningFeatureSelection _
) As MiningAttributeCollection
'Utilizzo
Dim instance As MiningModel
Dim filter As MiningFeatureSelection
Dim returnValue As MiningAttributeCollection
returnValue = instance.GetAttributes(filter)
public MiningAttributeCollection GetAttributes(
MiningFeatureSelection filter
)
public:
MiningAttributeCollection^ GetAttributes(
MiningFeatureSelection filter
)
member GetAttributes :
filter:MiningFeatureSelection -> MiningAttributeCollection
public function GetAttributes(
filter : MiningFeatureSelection
) : MiningAttributeCollection
Parametri
- filter
Tipo: Microsoft.AnalysisServices.AdomdClient. . :: . .MiningFeatureSelection
The MiningFeatureSelection value to filter on.
Valore restituito
Tipo: Microsoft.AnalysisServices.AdomdClient. . :: . .MiningAttributeCollection
A MiningAttributeCollection containing the attributes that have the specified feature selection value.
Osservazioni
When you create a mining model, Analysis Services chooses a subset of attributes to include in the data mining model based on statistical analysis. This attribute reduction process is called feature selection. The method for feature selection differs depending on the algorithm. For more information, see Funzionalità di selezione degli attributi nel data mining.
Feature selection can be applied to both input and output attributes. If you have many attributes, in a nested table, for example, it is possible that not all of the attributes will be included in the final model. Therefore, you can call the MiningModel method to determine which attributes are selected for input only, which attributes are selected for output only, and which attributes are selected for both output and input. You can also return all attributes or attributes that have no feature selection value.
Vedere anche