Here’s a natural English translation:
Hi, I’m new to KNIME and need to give a presentation on filter methods for feature selection. Is my workflow set up correctly, or do I need to apply the filters outside of the cross-validation? I’m feeling really unsure about this at the moment. Thanks a lot for your help!
I know this is a bit late for the original presentation, but for anyone finding this later: I suggest applying the same column selection to the test branch as well.
For example, you can use a Reference Column Filter and feed it the filtered training data as the reference. This ensures the test data has the same columns the model was trained on for each fold, preventing schema mismatches and giving more reliable accuracy results.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
