Dalex explainer made using testing or training data?

Hello,

I am going through the "Tidymodeling with R" book (https://www.tmwr.org/) and in chapter 18, I wondered why the explainer objects would be made with training data and not with testing data?

In the "Tidymodeling with R" book, explainer objects are made with training data instead of testing data because the goal is to explain the model's behavior based on the patterns learned during training. The testing data is meant to evaluate the model's generalization ability, not to inform the model's understanding. By using the training data for explainability, you can better understand how the model is making its predictions based on the features it was trained on, which is crucial for model interpretation.

Thank you! That helps a lot!

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.