Where to Start Learning Data Science – The most common challenge I see with people new to data science is figuring out where to start. There are countless different courses, certifications, degrees and bootcamps that you can take. Most people are too overwhelmed by all the options so they simply don’t pursue this awesome career path. In this article I give you my best tips on getting started in the field.
I bet you’re expecting me to give you the holy grail of learning. The step by step process for learning this field. Unfortunately, what works well for you may not work well for other people. For example, I know that formal education works really well for me, other people can self motivate and learn on their own far better than I can. The first part of learning is understanding yourself.
If you know the style of teaching that you like, it is a lot easier to find a starting point. To be completely honest, I think all of the courses out there are pretty good. The course that you take is less important than actually just getting started learning the material. If you get started and don’t like one of the courses, no one is preventing you from trying another one.
One of my subscribers found that he learned far more by doing his own projects than from following along with a course. This is totally fine, but he never would have known if he didn’t experiment with it. I personally like the free Kaggle micro courses, and I have made a video about the free resources that I like best. Again, all of these are good options and you can start anywhere.
I can’t stress this enough, just because you take one course, doesn’t mean that you can’t take another. This especially true for the free courses. You don’t just learn data science once, you are constantly reinforcing your knowledge and absorbing new resources. Don’t think of a single course as “how you are going to learn data science” it is just one of the many contributors to your data science body of knowledge.
If you think of it this way, starting is less scary because you aren’t missing out on anything, you are reserving time in the future to be able to check out other resources. The next thing that I would recommend when starting out is to immerse yourself in other people’s code. I would go on Kaggle.com and just look through the different notebooks there.
Kaggle is a place where people share the code that they used to analyze different datasets. You shouldn’t be discouraged if you don’t understand anything. This is an exercise in pattern matching. What tools do most people use? What visuals do you see a lot? What words are coming up repeatedly? You should take notes on what you see.
When you see these things coming up in a course you are taking or in other places, everything will start tying itself together. The next phase of learning is starting your own projects. In my opinion, starting a project is as simple as writing a single line of code. Sometimes it is enough to just do that. Just coding and going through examples and getting the code to run yourself counts as well.
After you get your code to run, you should experiment, what breaks it? Can you change your outputs? As you get more comfortable, you can start integrating the tools that you are using. Finally, you should begin to understand documentation. For all the packages you found in your notes, go through the documentation for them. This will be one of your most valuable skills as a data scientist. To learn even more about data science and to eventually become one, you basically just rinse and repeat these steps at a larger scale. Data science is a process or a journey that doesn’t necessarily have a destination.
I think that almost everyone is capable of learning this field if they get out of their own way.
reference – Where to Start Learning Data Science
Web enthusiast. Thinker. Evil coffeeaholic. Food specialist. Reader. Twitter fanatic. Music maven. AI and Machine Learning!