community
cancel
Showing results for 
Search instead for 
Did you mean: 

Alteryx designer Discussions

Find answers, ask questions, and share expertise about Alteryx Designer.
#SANTALYTICS

The highly anticipated Alteryx Community tradition is back! We hope you'll join us!

Learn More

Decision Tree: Error

Asteroid

I am running the decision tree model, but it always gives the error.

 

Decision Tree: Error: The minimum cross-validation error occurs for a CP value where there are no splits. Specify a complexity parameter.

 

Is there any documentation about how to set up the configuration? Thank you.

 

Capture.PNG

Alteryx Certified Partner
Alteryx Certified Partner

Hi,

 

What kind of dataset are you using? From what I can see the issue is, that the tool can't make any splits. What happens if you set the "minimum number of records needed to allow for a split" down to "1" (just for testing).

 

Further what happens if you go to "help->sample workflows->predictive Analytics->6 Decision tree" and run the sample? Does that work for you?

 

Best regards,

Daniel

I just ran into this same problem.  Turns out my target variable was set as a number field instead of a string.  Hope that helps 

Atom

After doing some troubleshooting, I'm pretty sure that my target variable was the reason why I got this error. Specifically, it was a binary categorical variable, and the Yes class constituted only 1.14% of 10,000 records. When I tried using a different binary categorical variable as the target -- one where the two classes were split 76/24 -- there was no error.

Hello Daniel, I have been facing the same issue and have tried all the procedure you suggested but still no positive result.

 

Thanks and Best Regards,

Fahad Munir

Labels