The SPARSEML Procedure
PROC SPARSEML Statement
PROC SPARSEML <options>;
The PROC SPARSEML statement invokes the procedure.
You can specify the following options:
- C=number
-
specifies the penalty value, where number must be a real number greater than 0.
By default, C=10.
- DATA=CAS-libref.data-table
-
names the input data table for PROC SPARSEML to use. The default is the most recently created data table. CAS-libref.data-table is a two-level name, where
- CAS-libref
refers to a collection of information that is defined in the LIBNAME statement and includes the
caslib, which includes a path to the data, and a session identifier, which defaults to the active session but which can be explicitly defined in the LIBNAME statement. For more information about CAS-libref, see the section Using CAS Sessions and CAS Engine Librefs.- data-table
specifies the name of the input data table.
- MAXITERS=number
-
specifies the maximum number of iterations before the process stops, where number is a positive integer.
By default, MAXITERS=100. In some cases, you can obtain a good model in fewer than 10 iterations.
- NTHREADS=number-of-threads
specifies the number of threads to use in the computation. The default value is the number of CPUs available on the machine.
- PRINTLEVEL=level-value
-
specifies a level value from which the ODS information tables are printed. You can specify the following level-values:
- AGGRESSIVE
generates all information tables.
- BASIC
generates basic information tables.
- MODERATE
generates standard information tables.
- NONE
suppresses information tables.
By default, PRINTLEVEL=MODERATE.
- REGL1=number
-
specifies the L1 norm penalization weight. If number is greater than 0, then the L1 norm penalization is performed.
By default, REGL1=0. If you specify a positive number, the penalty value C is ignored.
Note: The number must be greater than or equal to 0. You can specify either the REGL1= or REGL2= option, but not both.
- REGL2=number
-
specifies the L2 norm penalization weight. If number is greater than 0, then the L2 norm penalization is performed.
By default, REGL2=0. If you specify a positive number, the penalty value C is ignored.
Note: The number must be greater than or equal to 0. You can specify either the REGL2= or REGL1= option, but not both.
- SEED=value
-
specifies a non negative integer as the random number seed.
By default, SEED=1.
- TOLERANCE=number
-
specifies the minimal absolute tolerance at which the iteration stops. The number must be greater than 0.
By default, TOLERANCE=1.0E–6. In some cases, you can obtain a good model by using a number greater than or equal to 0.01.