Community Spring Cleaning week is here! Join your fellow Maveryx in digging through your old posts and marking comments on them as solved. Learn more here!

Alteryx Designer Desktop Discussions

Find answers, ask questions, and share expertise about Alteryx Designer Desktop and Intelligence Suite.
SOLVED

How to test a prediction model?

muhalim
7 - Meteor

Hi,

How can I test a model (e.g. Random Forest) generated from training dataset to a testing (or new) dataset correctly?

4 REPLIES 4
fmvizcaino
17 - Castor
17 - Castor

Hi @muhalim ,

 

You have 2 tools developed by Alteryx and you can download from Alteryx gallery.

You can connect your test set and your model and check the outputs from both tools to have a better idea of what model to use or how well your model is doing against your test data.

 

Cross validation

https://gallery.alteryx.com/#!app/Cross-Validation-Tool/5d9267cc8a933711e4473950

 

Model comparison tool

https://gallery.alteryx.com/#!app/Model-Comparison/5d92741a826fd30b84537a27

 

Best,

Fernando Vizcaino

huynv96
9 - Comet

You can use Score tool to test the model with new dataset

Capture.PNG

diksha1107
7 - Meteor

Hey @muhalim,

You can get the output of training or testing input data using the score tool as shown below. 

@huynv96  is also right but D input of score tool must be connected to Data and M input of score tool must be connected to O output of the Model.

 

Do let me know if you have any other doubt.

 

Warm Regards,

Diksha Adtani

 

 

score2.PNG

 

muhalim
7 - Meteor

Thank you all

Labels