Someone once told me than any time you answer a question more than once, you should just blog about it. To anyone just starting out in data science, here’s how I began my journey: like everyone else, I Googled it.
If you do, you’ll be presented with millions of resources, and I guarantee you won’t know where to start. The amount of available information is overwhelming.
I’d say that a background in statistics, mathematics, and/or computer science is important. I was fortunate to have done an undergraduate in Mathematics and Computers Science. However, these skills are easily accessible online. In this article, I’ll share some resources I’ve used.
I know there are so many free resources out there, but if you need guidance from the best professionals in the industry, you may have to pay for some of these courses. These websites usually offer coupons, and sometimes you can get a $200 course for $10. I’m not advertising any of these sites, but I am going to share what I’ve used.
- Sundog Education by Frank Kane. Frank spent 9 years at Amazon and IMDb, developing and managing technology that automatically delivers product and movie recommendations to hundreds of millions of customers. His course starts off with an introduction to statistics before diving into data science.
- Python for Data Science and Machine Learning Bootcamp. I really enjoyed this course. It’s very good if you don’t have a background in Python, because the instructor starts with Python basics that you’ll need for data science. If you’re already familiar with Python, you can just skip that part. The instructor, Jose Portilla, covers most of the things you need to know about data cleaning, visualization, and machine learning. He explains the concepts in ways that are very easy for any beginner to grasp.
- Deep Learning A-Z™: Hands-On Artificial Neural Networks. Kirill Eremenko explains the intuition behind every algorithm before Hadelin de Ponteves does the code exercises. I love their deep understanding of the concepts they’re teaching. The examples they use are also very relatable.
- Machine Learning A-Z™: Hands-On Python & R In Data Science. This also also by Kirill and Hadelin. Concepts here are covered in both Python and R. I don’t really like R so I skipped over the R tutorials. It’s a great one if you’re looking to learn Python and R at the same time.
- Intro to Machine Learning by Udacity. One the instructors of this course, Sebastian Thrun, is one of the co-founders of Udacity. He brings on board a lot of industry expertise. The course is free.
- Intro to Inferential Statistics
- Intro to Descriptive Statistics
- Springboard Learning Path Some good people at Springboard have curated most of the things you need to get started with data science.
- Kaggle Learning resources
- Kaggle Career Con 2018 videos. Earlier this year, Kaggle organized a digital conference event that focused on helping students and career switchers land their first data science job.
- Kaggle career con 2018 resources
- I have also used Kaggle Kernels and Kaggle competitions to see what other data scientist are doing.
- Blogs I follow: Springboard, Towards Data Science, Machine Learning Mastery, O’Reilly Media, No Free Hunch, Elite Data science, and Heartbeat
- For data sets, I use Kaggle and the UCI Machine Learning Repository.
I’ve also found that learning from other people can be a very good strategy. Here are a couple people I think would be valuable to follow because they blog about their data science experience; Rounak Banik, Ehi Aigiomawu, Derrick Mwiti, Catherine Gitau, Randy Lao, Kate Strachnyi and Siraj Raval.
I have also found official documentations of Python, Keras, Sklearn, Pandas, and all other packages used in data science to be very useful. Ultimately, you have to learn to use them.
Obviously this is still overwhelming for someone just starting out. I’d recommend beginning with the Udemy courses mentioned above. This is because they’re taught by professionals, and you can always ask them questions when you get stuck.
Otherwise, Towards Data Science and Heartbeat have great resources for you to get started for free. When you take these courses make sure you code along and do the exercises, otherwise you’ll have just wasted your time and resources. Without constant and consistent practice you can’t master these skills.
However, I must say it’s highly unlikely that you’ll become a pro data scientist without investing some dollars somewhere. So put in place a plan for that. Finally, as you learn, write articles about what you learn to encourage the next soul that tries their hand at data science.
Follow me on LinkedIn for more technical resources.
Discuss this post on Hacker News.