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=libref.data-table
-
names the input data table for PROC SPARSEML 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, and a session identifier, which defaults to the active session but which can be explicitly defined in the LIBNAME statement. For more information about 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 nonnegative 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.