I would like to know about if the node of Logistic Regression Predictor use only the coefficients that throws the Logistic Regression Learner node that area significant or have a P>|z| smaller than 0.05. This because I’m working with a confidence of 95%.
the Logistic Regression Predictor node uses all coefficients of the model.
If you want to exclude coefficients, I’d recommend to retrain your model without the features that aren’t significant.