OPINION
How to Self Study All The Technical Stuff You Need for Data Science
The 5 steps to master Python, SQL, statistics, and more.
There have never been so many free resources online to learn all the technical stuff you need for data science.
Do you want to learn Python? There are thousands of tutorials on Youtube. Do you want to practice SQL? You can google SQL exercises with solutions Do you want to learn statistics? Khan Academy has most lectures you need for data science
But when it comes to learning all of this on our own, things get a bit complicated. You might wonder where to start, how to study less and learn more, how to retain more information, how to reach your peak performance when studying, and how to make consistent progress.
In this article, I’ll show you the 5 steps I followed to learn the technical stuff needed for data science.
If you don’t feel like reading, you can watch my video instead!
Be sure to subscribe here to get my Python for Data Science Cheat Sheet I use in all my tutorials (Free PDF)
Step 1: Where to start? Build your own curriculum
Unless you were born yesterday, you already have some previous knowledge of data science (regardless of your background). You might already know some math and statistics or have used Excel or Python before, so you need to build a curriculum that best fits your needs.
Online courses’ curricula are fine, but they know little about you. This is why you have to pick a couple of them and customize your learning. Your own curriculum will help you focus on your weak points and skip those topics you’re already good at.
I remember that one of the first data science courses I joined had Excel lectures. I naively thought that there were “special Excel lessons for data science,” so I wasted 3 hours learning something I already learned at university (and even school!).
Don’t make the same mistake. Create your own curriculum and, if possible, a roadmap to know where you are now and what you’d like to know after 1, 3, and 6 months.
Remember that keeping track of your progress will help you stay motivated whenever your reach a plateau.
Note: Technical stuff like Python, SQL, and Machine Learning have applications that go beyond data science, so make sure you build your own curriculum based on well-known data science courses out there. Here you can find some courses I would take if I had to start over.
Step 2: Schedule tasks (Study less and learn more)
When it comes to acquiring technical skills, you have to be very careful with time!
I can’t count how many hours I wasted solving an issue that had little impact on my learning (like a silly error caused by a typo!). Things like that happen often, so you have to keep good track of your time by sticking to a schedule.
If you set 2 hours to learning a new Python library, then no matter how good or bad your day is, you’ll spend 2 hours wring Python code and then move on.
Having a schedule help you build the discipline you need for learning not only one but many things. We don’t always have control over how much progress we make on tool A, but we can make sure that if our learning gets slow, it doesn’t affect our progress with tool B.
How? Sticking to our schedule!
Consistency is crucial in the learning process. It’s not the same studying 7 hours every Monday to studying 1 hour from Monday to Sunday. The first will burn you out, but the second will give you the consistent practice necessary to learn new stuff.
Step 3: Learn the minimum stuff and start solving projects
When learning something new, we sometimes want to go deeper and understand why things work the way they do. Although this isn’t bad, when you want to learn technical things like a programming language, you need to learn the minimum stuff and put into practice what you’ve just learned.
Also, get some hands-on experience before moving to new topics. Keep learning new things is cool, but, first, we have to build a solid foundation on the previous topics.
How? Solving exercises and projects!
Say you’re learning how to use Pandas in Python. Instead of memorizing the hundreds of methods Pandas have, you should pick some of them and start solving a mini-project. Here are some ideas.
Pandas methods for cleaning data Pandas methods for making data visualizations Pandas methods for collecting data
Using these methods to solve meaningful tasks will help you easily remember their functionality.
Last but not least, make sure you repeat this process of learning/solving until you master a new topic (here’s yourfor loop!)
Step 4: Develop good habits to reach your peak performance
This is an important yet underrated point. Habits can boost your learning or slow you down.
Bad eating habits can make you feel so weak that you won’t be able to focus when studying. It can also stop your workflow every now and then or force you to pay a visit to a doctor whenever things get serious.
Bad sleeping habits will make your eyes close no matter how entertaining is the tutorial you’re watching.
In contrast, good habits will help you reach your peak performance and become a better learner.
Small habits like turning on Do Not Disturb mode every time you’re studying will help you avoid distractions and the habit of organizing your desk and computer will help you optimize your workflow and boost productivity, which leads us to the next point.
Note: To find a big list of habits that will make you a better data scientist check my articles here and here. There you’ll find habits to boost your technical, soft skills and life.
Step 5: Follow the 20-second rule to make consistent progress
This is a technique I learned from the YouTuber Thomas Frank many years ago. It’s based on a technique called “activation energy” explained in the book The Happiness Advantages.
In the video, Thomas explains that activation energy refers to the time, energy, or willpower to start doing something. The more activation energy an activity requires, the more willpower you need (this explains why it’s easier to play games on your phone than to start doing exercises)
The 20-second rule consists in lowering this activation energy, for those tasks we plan to do on a consistent basis. To do so, we have to strategically modify our environment to put hard activities on the path of least resistance.
Say your study room is filled with food and video games. The next time you get into your room, instead of studying, you’ll most likely eat something, grab your controller and start playing a video game.
Why? Such activities require less activation energy than studying.
But if you hide your controller and leave only things that you need for studying, you’ll increase the activation energy for playing video games, and decrease the activation energy for studying.
I took this technique to the next level by leaving only useful apps (iBooks, audible, notion, fitness+, etc) on the home screen and hiding distracting apps (social media and games) on folders located on the last page.
The result? Every time I unlock my phone I remember I need to read books, check my notes and do physical activity.
Bonus: Have enough breaks — Here’s where the best ideas come from
Sometimes we push ourselves too much and forget about having those breaks that can help us boost our learning.
My best ideas come when I’m taking a walk, resting, or having a shower. Even Elon Musk attributes cold showers (one of his daily habits) to the source of many ideas.
I also found out that whenever I have a fresh mind, I’m able to connect the dots of what I’ve learned. That is, finding the connection between topic A and topic B and being able to put them together, say, in a project.
Breaks are important, so have them regularly.
If you enjoy reading stories like these and want to support me as a writer, consider signing up to become a Medium member. It’s $5 a month, giving you unlimited access to thousands of Python guides and Data science articles. If you sign up using my link, I’ll earn a small commission with no extra cost to you.