BookmarkSubscribeRSS Feed
tom_evans_79
Fluorite | Level 6

Hi,

 

In Enterprise Miner there was an 'Append' node that joined together two tables or subtables. Is there an equivalent in Model Studio?

 

In some situations, it is better to build two independent models to be applied to different parts of a population. To find the best way to split the sample, we want to build models with various splits and compare the discriminatory power of the combined results. Model Studio is great for comparing different models, but to do this we need to be able to split and then recombine different parts of the sample. I think I can do the splitting part in Model Studio, but can't find a way to re-attach the two parts such that each part of the split is scored by the correct model. Any ideas?

 

Here's the Enterprise Miner flow that we want to replicate in Model Studio:

EM Diag.PNG

 

 

 

 

 

 

 

Many thanks,

 

Tom.

2 REPLIES 2
gcjfernandez
SAS Employee
Please use the SAS code node to append the two scored data files.
tom_evans_79
Fluorite | Level 6

Thanks for the reply, @gcjfernandez .

 

I can only connect 1 predecessor node to a SAS code node, but for the node to work as a generic 'append' it would need at least 2 predecessors.

 

I could probably work out how to merge the two tables using SAS code, but if the code would need changing every time then it isn't worth using DMML.

 

Tom.

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
  • 2 replies
  • 2283 views
  • 0 likes
  • 2 in conversation