Using a Framework for Data Manipulation

An applied ML solution is just for that specific use case. It's usually not meant to generalize. And frameworks are meant to support many purposes in an abstract way. So if we're building this particular model and we're trying to classify whether our users are committing fraud or not, well, there's going to be specific data manipulations that we're going to do. Whereas in a framework, you don't build stuff like that into a framework.