KPCA Procedure

SAVESTATE Statement

SAVESTATE RSTORE=libref.data-table <option>;

The SAVESTATE statement creates an analytic store for the model and saves it as a binary object in a data table. You can use the analytic store in the ASTORE procedure to score new data. For more information, see Chapter 4, ASTORE Procedure.

You must specify the following option:

RSTORE=libref.data-table: specifies the name of the table in which to save the analytic store. libref.data-table is a two-level name, where libref refers to the library, and data-table specifies the name of the output data table. For more information about this two-level name, see the DATA= option and the section Using SAS Viya Workbench.

You can also specify the following option:

EXACTSCORE: specifies the exact method of scoring. If this option is not specified and METHOD=APPROXIMATE in the PROC KPCA statement, then fast (approximate) scoring is used. However, if this option is not specified and METHOD=EXACT in the PROC KPCA statement, then exact scoring is used. For more information about fast scoring, see the section Fast Scoring of KPCA.

You can use the ASTORE procedure to specify the number of principal components to project the scoring observations. To do this, specify the option KPCA_NPC in the SETOPTION statement in PROC ASTORE. If you specify the PREIMAGE statement, you can use PROC ASTORE to obtain the pre-image of the data in the original input space by specifying the value of SCORING_MODE as 1 in the SETOPTION statement in PROC ASTORE. You can also specify other pre-image scoring options, including ITER_NPC, ITER_NUM, and ITER_THRESH, in the SETOPTION statement in PROC ASTORE. For more information, see the sections KPCA Iterative Pre-image Method and KPCA Mapping Pre-image Method.

Last updated: September 23, 2026