BookmarkSubscribeRSS Feed
🔒 This topic is solved and locked. Need further help from the community? Please sign in and ask a new question.
Mats
Fluorite | Level 6

I'm looking at the SAS Viya machine learing demo. It races some machine Learning algorithms against each other on a given dataset. All models produce almost equally good "lift" as shown in lift diagrams in the output.

 

If you tweak the Learning to perform on a smaller subset of the data; only 0.002% of the total data set (proc partition data=&casdata partition samppct=0.002;), most algorithms get into problems producing lift.

 

But the neural network is still performing very well. Feature or bug? I could imagine that the script does not re-initilize the network, but it is hard to guess from the calls alone.

1 ACCEPTED SOLUTION

Accepted Solutions
BrettWujek
SAS Employee

Mats - the short answer without running some studies of my own is that neural networks are highly adaptive and can train very accurate models with far fewer observations than many other techniques.  The tree-based models are going to be quite unstable with very few observations.  In this case you sampled all the way down to around 20 observations...even that might be sufficient for a neural network if the space it not overly nonlinear.

 

As for your last comment - it seems you are referring to what is known as warm start, where a previously trained model can be used as a starting point and refined by providing new observations.  That is NOT what is happening here, as that capability is only coming available in our upcoming release which is just over a month away.

 

Thanks for trying out the software and experimenting...please continue to provide feedback and ask questions over this forum.

 

Brett


Register today and join us virtually on June 16!
sasglobalforum.com | #SASGF

View now: on-demand content for SAS users

View solution in original post

6 REPLIES 6
BrettWujek
SAS Employee

Mats - the short answer without running some studies of my own is that neural networks are highly adaptive and can train very accurate models with far fewer observations than many other techniques.  The tree-based models are going to be quite unstable with very few observations.  In this case you sampled all the way down to around 20 observations...even that might be sufficient for a neural network if the space it not overly nonlinear.

 

As for your last comment - it seems you are referring to what is known as warm start, where a previously trained model can be used as a starting point and refined by providing new observations.  That is NOT what is happening here, as that capability is only coming available in our upcoming release which is just over a month away.

 

Thanks for trying out the software and experimenting...please continue to provide feedback and ask questions over this forum.

 

Brett


Register today and join us virtually on June 16!
sasglobalforum.com | #SASGF

View now: on-demand content for SAS users

Mats
Fluorite | Level 6

Thanks a lot for the fast and clear answer. I also posted my quesion on Stack Overflow. Is it OK if I post your answer there as well? /Mats

Mats
Fluorite | Level 6
Thanks a lot for the fast and clear answer! I also posted my quesion on Stack Overflow. Is it OK if I post your answer there as well!? /Mats
BrettWujek
SAS Employee

Hi Mats - Certainly feel free to post my answer on Stack Overflow.  If you don't mind, would you provide the link to that posting here so that people can jump over and see what others might have to say.  Cross-posting is always good :-).

 

Thanks.


Register today and join us virtually on June 16!
sasglobalforum.com | #SASGF

View now: on-demand content for SAS users

Xinmin
SAS Employee

Mats, PROC NNET initializes weight random, if you specify a seed in the train statement, the initial weights are repeatable. NNET training is powered by a sophiscated nonlinear optimization solver, if the log shows "converged" status, it means the model is fit very well. 

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
  • 6 replies
  • 1536 views
  • 4 likes
  • 3 in conversation