Automodel learn/test

DocMusher
DocMusher New Altair Community Member
edited November 5 in Community Q&A

Is there a good reason to split the data in 60/20/20% where the last 20% is used to "test the testing of the conclusion", as proposed in another platform?

Tagged:

Best Answer

  • varunm1
    varunm1 New Altair Community Member
    edited June 2019 Answer ✓
    Hello @DocMusher

    Looks similar to RM automodel, where the data is split into 60:40 (train: test) but the 40% test data is again split into 7 hold out sets to test the model.

    We split the 40% again into 7 parts, evaluate the model on each part, get rid of the two extremes/outliers, and build the average of the rest.  This way we keep many of the benefits of a cross-validation without it's biggest drawback: 5x-10x runtime increases. (Explanation from @IngoRM)
    @IngoRM might add more here.

Answers

  • varunm1
    varunm1 New Altair Community Member
    edited June 2019 Answer ✓
    Hello @DocMusher

    Looks similar to RM automodel, where the data is split into 60:40 (train: test) but the 40% test data is again split into 7 hold out sets to test the model.

    We split the 40% again into 7 parts, evaluate the model on each part, get rid of the two extremes/outliers, and build the average of the rest.  This way we keep many of the benefits of a cross-validation without it's biggest drawback: 5x-10x runtime increases. (Explanation from @IngoRM)
    @IngoRM might add more here.

Categories