Resolved: StandardScaler of features

The main idea is to align the train data distribution with the one of the test data; the transformation of both the train and test data must be consistent. Otherwise, if you only standardize the training data and you don't standardize the test data, you won't be able to generalize, i.e. your model will not perform well when applied to unseen data, wihch is the primary goal of creaeting an ML model.