A big trap in data science education is:
- learning DS libraries before learning coding basics
- learning ML algorithms before learning how to preprocess your data
- learning Deep Learning before Machine Learning
- Learning data viz before understanding the basics of statistical inference
This is where most people may spend months trying to read a book, finish a MOOC, or learn a topic on YouTube with little retention of the applied topic.
Anytime you’re learning a new concept you can’t neglect the fundamentals & always keep the big picture in mind because:
You have to know coding basics before you can even debug the implementations of your DS/ML Libraries.
You have to know how to preprocess your data before applying machinelearning methods accurately.
You have to know statistical inference before you make sense of your visualization.
Remember to don’t just jump into a course without taking the time to ask yourself “why” and “how” is this being used.
It’s only when you start asking yourself questions about the problems where you begin to connect the dots more.
And by having a good foundation of the basics, those dots that you connect with new concepts will be retained a lot longer.