There's a question in her comments:

This looks like a significant advantage of MLForecast! Does the integrated preprocessing offer any flexibility for customizing the preprocessing steps, or is it focused on a specific set of transformations?

Could anyone reply, or help me with a response? Thank you!