BookmarkSubscribeRSS Feed
autoleech
Calcite | Level 5

Hi all,

I started using SAS EM not too long ago and I am having a very basic issue. I have a data set that when I do not specify a target variable, StatExplore runs fine. When I do specify my target variable, StatExplore gives me an error and asks me to consult the log. I do not understand what the log is saying, but I don't see anything jumping out at me. Something that is odd about this, is that I have the same data set with one less column in it, and when I specify a target in that data set StatExplore runs fine. I head back to the other data set and reject the only variable creating a difference between the data sets and it still gives me an error.

Are there are any basic steps I can take to begin solving this issue?

Thanks in advance.

6 REPLIES 6
mohamed_zaki
Barite | Level 11

what is the table role of your data set as it appear in the Properties Panel when you select your data node?

Also how do you use the statexplore node and what kind of information you need from it?

autoleech
Calcite | Level 5

It is set to RAW.

mohamed_zaki
Barite | Level 11

Also how do you use the statexplore node and what kind of information you need from it?

How many variables do you have what is the roles assigned to them?

M_Maldonado
Barite | Level 11

autoleech,

I had no luck at trying to reproduce this weird error.

To expedite things let's do this:

1. submit a tech support form here Contact SAS Institute Technical Support

2. if your data is not confidential, please attach a tiny sample (csv,txt,sas, any format) and a screenshot of your metadata, and also confirm your EM version. I will give it a try later today.

If it is easier to deal via email my contact is at the end of this link: http://support.sas.com/resources/papers/proceedings14/SAS133-2014.pdf

Thanks,

Miguel

autoleech
Calcite | Level 5

Thanks for the replies and sorry for my tardiness in responding.

I can run my model without StatExplore, so I will do that for the moment. I was hoping to see some correlation information with StatExplore.

I will follow MiguelMaldonado's advice if StatExplore becomes necessary, although the data is confidential so I'm not certain how far we can go.

autoleech
Calcite | Level 5

I am not sure if I should create a new thread for this, but I have another question.

I would like to build a model in SAS EM. Is it possible to output this model to use in SAS EG? For example, say that I want to use linear regression. Is there some way that SAS EM outputs the function y^ = a_1 x_1 + ... + a_n x_n?

I am really interested in the y^ produced by a gradient boosting model.

If it is not possible for SAS EM to produce this function y^, is it possible for me to place a model created in SAS EG into SAS EM for comparison?


sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

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
  • 6 replies
  • 1632 views
  • 6 likes
  • 3 in conversation