Specifies the parameters used by the Analysis Services server when it performs cross-validation of mining models.
These parameters are for internal use only. The information provided here is for reference only.
<MiningModel> ... <ddl100_100:FoldingParameters>... <ddl100_100:FoldIndex>...</ddl100_100:FoldIndex> <ddl100_100:FoldCount>...</ddl100_100:FoldCount> <ddl100_100:MaxCases>...</ddl100_100:MAxCases> <ddl100_100:FoldTargetAttribute>...</ddl100_100:FoldTargetAttribute ...</ddl100_100:FoldingParameters> </MiningStructure>
|FoldIndex||Integer that indicates the starting position of the partition that is used for cross-validation.|
|FoldCount||Integer that indicates the number of partitions in the model after cross-validation.|
|MaxCases||Integer that indicates how many model cases are used for cross-validation.
A value of 0 indicates that all cases are used.
|FoldTargetAttribute||String that indicates the ID of the model column that contains the predictable attribute.|
These properties are for internal use only, and are not supported for use in DDL statements.
For information about how to use cross-validation in SQL Server Data Tools (SSDT), see Measures in the Cross-Validation Report.
For information about how to perform cross-validation by using Analysis Services stored procedures, see Data Mining Stored Procedures (Analysis Services - Data Mining).