"Is it possible to obtain the predictions made inside an operator of X-Validation"

earmijoearmijo Member Posts: 270 Unicorn
edited June 2019 in Help
In each round of a 10-fold X-validation process, one tenth of the dataset is used as a test set and the rest is used as training data. The model is used to predict that one tenth. After 10 rounds you get a full vector of "conservative" predictions. These predictions are compared to the actual values and metrics like the RMSE are calculated.

My question is: Is there a way to obtain those predictions inside the X-validation operator?


  • Options
    awchisholmawchisholm RapidMiner Certified Expert, Member Posts: 458 Unicorn

    You could store the results in the repository using the Store operator or in memory using the Remember operator. To avoid overwriting, you would have to make sure the name of the stored item was changed based on the value of a macro that incremented as each iteration of the loop happens.

    Outside the cross validation, you could retrieve or recall as needed.


  • Options
    earmijoearmijo Member Posts: 270 Unicorn
    Thank you very much Andrew. I'll give it a try.
  • Options
    earmijoearmijo Member Posts: 270 Unicorn
    Oops. This is embarrasing...

    I just discovered there is an operator in RM that does precisely what I want: X-Prediction.
Sign In or Register to comment.