Configuring the Algorithm

Configure SVD for optimal performance, model size, and projection methods, including PCA.

Several options are available for configuring the Singular Value Decomposition (SVD) algorithm. Among several options are: settings to control model size and performance, and whether to score with SVD projections or Principal Component Analysis (PCA) projections.

See Also:

DBMS_DATA_MINING — Algorithm Constants and Settings: Singular Value Decomposition for a listing and explanation of the available model settings.

Note:

The term hyperparameter is also interchangeably used for model setting.