I have trained few models using sklearn and python. However I have scaled the data for Support Vector Machines - standarized and for Neural Network - Scaled [0-1] since it gives me better results and I believe it should be like that.
Since the ensembles usually gives better results than weak classifiers I would like to take advantage of that. I can train the SVM on standarized data and that's fine. But when i feed ensemble, let's say Stacking or Voting I'm just passing the non-scaled data because there are many other models and the models do not need scaled data.
I have struggled using pipeline. How do you handle such a problem?