SVDD Procedure
PROC SVDD Statement
PROC SVDD <options>;
The PROC SVDD statement invokes the procedure. You can specify the following options:
- DATA=libref.data-table
-
names the input data table for PROC SVDD to use. The default is the most recently created data table. libref.data-table is a two-level name, where
- libref
refers to a collection of information that is defined in the LIBNAME statement and includes the
library, which includes a path to the data. For more information about libref, see the section Using SAS Viya Workbench.- data-table
specifies the name of the input data table.
- NBINS=number
specifies the number of bins to display, where number must be an integer greater than 2. By default, NBINS=20.
- NOPRINT
suppresses the generation of ODS outputs. If you specify this option, no ODS tables are generated.
- NTHREADS=number-of-threads
specifies the number of threads to be used in the computation. The default value is the number of CPUs available in the machine.
-
OUTLIER_FRACTION=number
FRACTION=number
FRAC=number -
specifies the expected fraction of the training data that consists of outliers, where number is a real number between 0 and 1.
By default, OUTLIER_FRACTION=1E–6.
- SCORE
scores the observations in the data table that is specified in the DATA= option and generates a frequency table of
_SVDDDISTANCE_values.-
STD
STANDARDIZE standardizes the variables that are specified in the INPUT statement with LEVEL=INTERVAL by subtracting their mean and dividing by their standard deviation.
-
UCC
USECONSTANTCOL uses variables that have constant values for SVDD scoring. (By default, the variables that have constant values are ignored.) This option is useful in scenarios where certain variables have constant value in the training data, but variation in the values of those variables is expected in the scoring data. This option ensures that variables with constant values are used for scoring.
-
OUTSV=libref.data-table
SV=libref.data-table specifies the output data to contain the training data observations that are identified as support vectors, the corresponding value of the Lagrange multiplier (column
_SVDDALPHA_), and an indicator (column_SVDDOUTLIER_) that flags the position of the support vectors relative to the data boundary. Support vectors whose_SVDDOUTLIER_value is 0 fall on the data boundary, and support vectors whose_SVDDOUTLIER_value is 1 fall outside the data boundary. 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.