BookmarkSubscribeRSS Feed
MLAC
Fluorite | Level 6

I am using Enterprise Miner to train a Logistic Regression Model. In my dataset, there is a nominal attribute which has 513 classes. So, the training was not performed since it exceeded 512. Then, I followed the instruction to change the EM_TRAIN_MAXLEVELS to 550, I also added "%let EM_TRAIN_MAXLEVELS = 550" in Project Start Code. I used a new Regression node to run the flow, but it still didn't work. This time the error is "Error: Maximum target levels of 550 exceeded." 

 

What am I supposed to do? 

1 REPLY 1
Reeza
Super User
Are you sure your nominal attribute has 513 levels? If it did, 550 levels would have been enough.
Verify the number of levels within then nominal variable. You can use the following code or a proc freq to do that.

proc sql;
select count(distinct varName) as N_Distinct
from tableName;
quit;

sas-innovate-2024.png

Available on demand!

Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.

 

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
  • 1 reply
  • 1586 views
  • 0 likes
  • 2 in conversation