The second step of ETL pipelines is transform. Transformations can be split into two types—those that affect data samples individually, and those that affect a dataset as a whole. In the following paragraphs, we will cover the former transformations and explain how our samples can be preprocessed.




















































