The internet has thousands of posts on learning Data Science (DS), but rarely do you find any on how to fill in the holes. I am sharing my approach with those that are interested. I am sure others have their systems, and I would love it if you would share them.

My Approach

• I am a big champion of creating passion projects for your portfolio. I quickly learn what I don't know when I reach a roadblock. I will then Google and check the resources below to learn more. Some projects are big, some small. I find happiness when I hit a roadblock later in the process. To me, it means I am getting better.

• Sometimes, the project is to learn a specific topic. I need to determine what level am I at - Know Nothing, Beginner, Mastery, or World Expert? Then I challenge myself to get to the next level. I may only try to jump up 1 level of knowledge or do a deep dive to learn as much as possible. I will freely admit I am far from becoming a world expert, but I will get there. I am currently diving into Data Wrangling and Exploratory Data Analysis. I know enough to get by, but I am looking for my holes. For this project, my finished "project," I have assigned myself to write at least one blog post and to create a code cookbook for myself on the topic. Since this is a deep dive, it will take some time. I may do multiple posting along the way since the scope of the subject is enormous if the goal is a world expert. The end goal of becoming a world expert is a big one; I will permit myself to do small side projects alongside this deep dive. I might pause the topic once I level up and go back later. I do give myself some flexibility in my learning journey.

• I give myself homework. With any "Live/Real" class, you have to do something (assignment, project, test) to show you have learned and understood the material. You can only read so many blogs or watch so many videos. If you don't do it yourself - you don't own the material. I require deliverables. Some deliverables I keep private. I am considering a leap to make all my deliverables public. There are pros and cons to having everything public.

• Give yourself a deadline and assignment/goals. Treat your learning like a class. Deadline - firm hard deadlines, accelerate your understanding by requiring you to work on your plan regularly. I also understand where I am at and what I can do. If work has a huge deadline, I will not have a passion project deadline in the same week.

• A class with a good foundation/overview (See resources below) provides a good outline. From there, you take each step and how do I get mastery of that topic. A course is never going to make you a master's level. To master something, you have to do and keep doing. This is precisely why Ph.D. programs usually require a thesis. You need to do. This is also why participating in Kaggle Competitions can also help you improve.

• Review. Have an honest self-review. Where are my strengths? Where are my weaknesses? Then ask what the one thing I can do to level up one strength and one weakness is? Then take those steps. Repeat this review regularly.

What I have accepted

With any system, there are limitations; here is what I have accepted as the limitations of my approach.

1. First, I accept that I will always miss something no matter how good I am. There are always ways to improve myself. If I am always working to improve, I will keep getting better.

2. I have a target where do I want to be that is specific. For example, I want to work in Science, specifically Cancer Research, Life Science, or Ocean Research. By doing projects in these areas, I am building up a skill that meets my end goal. Yes, this means I might not be learning something that would help me in, let's say, Finance Tech, and that is ok for me. Do I learn many things that would crossover - yes? I know I cannot do everything, so I focus on where I want to be / what I want to do. Later if I change the course, I still have a good foundation. I will also do some projects just for fun, but I know my destination. I work towards my goal.

Resources I Check

There are many roadmaps out there and valuable tools. Here are a few I check.

• Open Source Learning Path

  • These sites have created a path to learning Data Science (DS). I will be honest I don't follow them from start to finish, but I do visit when I want to deep dive into an area. These links are the ones I check out first.

  • [Open Source Masters](http://datasciencemasters.org/)

    • The Curriculum for Data Science creating a learning path for all steps in the learning pathway with free videos/blogs. Sometimes they will link a book that has a small cost.

  • [Open Source Society University](https://github.com/ossu/data-science)

    • Useful resource covering all the key topics.

  • [Data Science Learning Roadmap for 2021](https://www.freecodecamp.org/news/data-science-learning-roadmap/amp/)

    • This is an excellent resource with links to free content plus estimated time to spend on each topic.

• [Python for Everyone](https://www.py4e.com/) is a great way to learn python.

• Kaggle

  • [Kaggle Courses](https://www.kaggle.com/learn)

  • There are tons of how to become / how to learn /what to do next to learn Data Science on Kaggle. Here is just one example that provides an excellent checklist

    • [How to Become a Data Scientist at Your Own](https://www.kaggle.com/getting-started/44915) by Zusmani

• Universities

  • There are a few open master programs for data science see above. I know what topics are needed for that area. I also look at Master Programs at Ivy League and top-tier schools. I know what they require for their students. Look at the class requirements for a master's in the US at Stanford and MIT. Georgia Tech has a highly rated and relatively inexpensive online Master's program. Is self-study the same as a top-tier class? Probably not, but there is some overlap, and the cost-saving is significant.

• Certifications

  • I also look to see if there are any certifications. I might not take the exam, but I will review the material to see if I know it. If I don't know something, I will look for free study materials for what I miss.

Good luck with your Data Science Journey. If you have any questions, let me know. If you have your system for ensuring you don’t have holes in your learning, I would love to learn about it.