pipelines chains together multiple steps so that the output of each step is used as input to the next step
pipelines make it easy to apply the same preprocessing to train and test!
and if you want to implement a classifier like DecisionTreeClassifier or RandomForestClassifier call (.fit) not (.fit_transform)
if you do not want to use the classifier then (.fit_transform) will perform great
credits:https://www.youtube.com/watch?v=xOccYkgRV4Q&list=PLKnIA16_Rmvbr7zKYQuBfsVkjoLcJgxHH&index=36