Great Article Snigdha. I enjoyed it.

i do have a question, why do we retrain again on both the test and training set, shouldn’t the model be taken as out of sample where the training set finishes? how else will we be able to validate the model out of sample?

thanks.