The HPNEURAL Procedure

The primary purpose for training a neural network model is to use the trained model to score new data that was not seen during training. It is very important that the new data have the same statistical characteristics as the data that were used for training. The following statements use the model that was trained in the preceding section to score some new observations:

title 'New Iris Data';

data new_iris;
   input SepalLength SepalWidth PetalLength PetalWidth;
   datalines;
50 33 14 02
64 28 56 22
65 28 46 15
67 31 56 24
63 28 51 15
46 34 14 03
69 31 51 23
62 22 45 15
59 32 48 18
46 36 10 02
61 30 46 14
60 27 51 16
65 30 52 20
56 25 39 11
65 30 55 18
58 27 51 19
;

proc hpneural data=new_iris;
   score model=model_iris out=scores_new_iris;
run;