Feature selection

Hi, im kevin and im new in knime, i need to know feature importance score for my regression. The output is double and I cannot use double as a target column in global feature importance node. What should I do to have feature importance score (percentage) for my regression (the output is double). Thank you

Iā€™m told these some of these extensions also work with Regression problems:

Here are examples for these extensions:

Global Feature Importance at the moment only works for classification problems.

1 Like

Hello Kevin,

I would suggest using the XGBoost Tree Ensemble Learner (Regression) node for your Regression task. This node outputs the feature importance table along with the Regression Model.

Please find below the KNIME Hub link of this node

Let me know if this works for you.


This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.