Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Enterprise Miner's IGN node is probably the most frustrating node ever

Reply
Occasional Contributor
Posts: 13

Enterprise Miner's IGN node is probably the most frustrating node ever

I don't know if there are other analysts out there who struggle as much with Enterprise Miner's Interactive Grouping Node as much as we do. My team and I need to use this node regularly in E-Miner (version 13.2) in order to conduct our work, yet this thing breaks so much and gives errors on a regular basis - which is really frustrating.

 

Currently, I need to use the node with a special codes dataset. The special codes dataset I import in E-miner have the exact same format as previous special codes datasets that I have used with great success (and it also has the same format as in E-miner's help file). When I run the IGN node with the special codes dataset and using the Quantile binning method (which is the method I prefer), the node stops and gives me an error. When I inspect the log, I find an error stating "Error: there was one unclosed do block". How do you even fix that if it is not you coding?? Anyway, when I run the node without the special codes dataset, it works fine. On the opposite side, when I run the node along with the special codes dataset and select the Bucket binning method instead, it also works fine. So, from what I understand, using the combination of a special codes dataset and the Quantile binning method results in this thing giving an error. And it is not as if E-miner's log-file is easy to inspect so that I can determine what the error is.

 

Has anyone experienced similar problems and/or found ways to fix it?

 

Thanks

J

Super User
Posts: 7,859

Re: Enterprise Miner's IGN node is probably the most frustrating node ever

Posted in reply to JakesVenter

What does SAS Technical Support say?

---------------------------------------------------------------------------------------------
Maxims of Maximally Efficient SAS Programmers
Ask a Question
Discussion stats
  • 1 reply
  • 184 views
  • 0 likes
  • 2 in conversation