The NNET Procedure
Syntax: NNET Procedure
The following statements are available in the NNET procedure:
PROC NNET <options>;
INPUT variables </ LEVEL=INTERVAL NOMINAL>;
HIDDEN number </options>;
TARGET variables </options>;
TRAIN OUTMODEL=CAS-libref.data-table <options>;
ARCHITECTURE architecture-options;
WEIGHT variable;
PARTITION <partition-options>;
OPTIMIZATION <options>;
AUTOTUNE <options>;
CROSSVALIDATION <options>;
OUTPUT OUT=CAS-libref.data-table <option>;
CODE <options>;
When you train a neural network, the PROC NNET, INPUT, TARGET, and TRAIN statements are required. At least one HIDDEN statement is required unless you use the GLIM architecture; in that case the HIDDEN statement is not allowed.
When you use a previously trained neural network to score a data table, only the PROC NNET, OUTPUT, and CODE statements are allowed.