BookmarkSubscribeRSS Feed
deleted_user
Not applicable
Hi All,

Kindly help me out. I have a model which is based on decision trees in Enterprise Miner and I need to score a new dataset. Since the final (tree) model selects only 4 out of 10 variables so I have imported a SAS dataset that has these 4 variables. In order to find accuracy of the final model I have randomly selected data from original SAS dataset and connected the Assessment node and the new dataset to be scored with the Score node. In the new dataset to be scored there are some missing values.

When I run the Score node, I get error:

27838 data EMDATA.VD_XXW9B /view=EMDATA.VD_XXW9B;
27839 set EMDATA.SVAL3155;
27840 run;

NOTE: DATA STEP view saved on file EMDATA.VD_XXW9B.
NOTE: A stored DATA STEP view cannot run under a different operating system.
NOTE: View EMDATA.SVAL3155.VIEW used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds



ERROR: Scoring data set is not identified.

27841 proc printto;
27842 run;


The Run Score[Apply] fails.

Kindly suggest.
4 REPLIES 4
SlutskyFan
Obsidian | Level 7
I'm actually new to EM, so this may not be good info. I don't understand the errors you are getting-but, I wonder, are you also using a SAS code node in the schema as well?

Is your score data in a library that is connected to Enterprise Miner?

These may be basic questions that you are far past, but I am a novice with EM. If you are not using a SAS code node in your flow- see here for an example:

http://econometricsense.blogspot.com/2011/02/scoring-data-in-sas-enterprise-miner.html
SlutskyFan
Obsidian | Level 7
If you checked my previous link earlier, I forgot to add the model node- sorry. It has been corrected. You can see the correct version here: http://econometricsense.blogspot.com/2011/02/scoring-data-in-sas-enterprise-miner.html
David_Duling
SAS Employee
In addition, this kind of problem is perfect for tech support since we cannot really see the details of what was done here in this forum. In some way, the view path to the data set was not resolved, but we don't know any more than that.
SatishG
Calcite | Level 5
Did you change the Role of the dataset you created for Scoring to 'Score'? When you create a data source, the default Role is 'Raw' which means the dataset can be used for Model Training. And for scoring it has to be 'Score'.

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 4 replies
  • 1551 views
  • 0 likes
  • 4 in conversation