In the notes for Module 4, section “Advanced Predictive Modeling”, chapter 7, “The AutoNeural Node”, page 7-4, the author states that “The AutoNeural Node can be used to automatically generate a neural network. The code that it generates can be used as a starting ...”.
How can a user get access to the code that the AutoNeural Node generates?
Thank you,
Bill Donaldson
SAS Enterprise Miner (EM) nodes generate two types of code: (1) train code, and (2) score code. Train code is the code used by EM to fit a model or otherwise produce the results and exported data created by the node. To get train code, open the results window, then select View->Properties->Train Code. The train code usually is dominated by SAS macro calls, so it may be impossible to see exactly how results are produced. Score code is the code used to score data sets having the same variables used to train a model. To get score code, open the results window, and then select View->Scoring->SAS Code. EM supports two ways of scoring. The AutoNeural node produces score code that can be included in a data step, which can be characterized as "data step score code." Data step score code can only be successfully deployed inside a SAS data step. There is a second type of score code that can be characterized as "not data step score code." The second type of score code may require execution of one or more SAS procedures and additional code that is executed as stand alone code. For example, forests and support vector machines have separate procedures for efficient scoring.
SAS Enterprise Miner (EM) nodes generate two types of code: (1) train code, and (2) score code. Train code is the code used by EM to fit a model or otherwise produce the results and exported data created by the node. To get train code, open the results window, then select View->Properties->Train Code. The train code usually is dominated by SAS macro calls, so it may be impossible to see exactly how results are produced. Score code is the code used to score data sets having the same variables used to train a model. To get score code, open the results window, and then select View->Scoring->SAS Code. EM supports two ways of scoring. The AutoNeural node produces score code that can be included in a data step, which can be characterized as "data step score code." Data step score code can only be successfully deployed inside a SAS data step. There is a second type of score code that can be characterized as "not data step score code." The second type of score code may require execution of one or more SAS procedures and additional code that is executed as stand alone code. For example, forests and support vector machines have separate procedures for efficient scoring.
Thank you for the pointer.
This is a knowledge-sharing community for learners in the Academy. Find answers to your questions or post here for a reply.
To ensure your success, use these getting-started resources:
Estimating Your Study Time
Reserving Software Lab Time
Most Commonly Asked Questions
Troubleshooting Your SAS-Hadoop Training Environment
Ready to level-up your skills? Choose your own adventure.