Test

This tab is display for Classification, Regression, and Anomaly Detection models; they are the only model can be tested.

The test tab defines how tests are done. The default is to test all models built. By default, the test data is created by randomly splitting the build data into two subsets; 40% if the input data is used for the test set.

You can specify no test. If you do perform a test, you can use all of the build data for test, modify the split ratio, or use a a test source for test.

Another way to test a model is to use a Test Node.