Learning data science? Steal this data scientist’s list of resources to get started

Photo credit: Thought Catalog.

Many people have asked me how to improve or even start with data science. Coming from an economics and accounting background, I had limited exposure to coding/programming until midway of taking my master’s degree six years ago (apart from some very basic introduction to programming in high school and Excel operations/functions). So, getting into data science for me meant going through an exploration phase, picking up different skills to fill the role.

This post is meant to explain what worked for me and only focuses on the technical bits. This is because there are qualitative attributes that help a lot in data science—formulating data science problems, articulating results, making actionable analyses, etc.—that come inevitably with experience in corporate environments.

As this is just my personal opinion and experience of a good pipeline, consider other options/ideas too. This pipeline assumes you have no programming or direct machine learning/data science experience. Some high school-level math is required.

Learn the basics of a programming language

The first thing you need is to learn Python or R (I prefer Python and am not the best person to advise on the latter as I rarely use it).

This is an easy-to-read ebook for Python that explains the basic elements like loops and if statements, variables, and lists. There are good resources in Python’s website and other books. There’s also an interactive tutorial.

I’ve personally used only the first ebook, which was enough for me to move on. Occasionally, I went back to different sources to learn specific concepts. However, after you understand the concept, StackOverflow is your best friend.

Set up data science packages

Once you learn the basics of the programming language of choice, you need to install the necessary packages that will allow you to exploit different data science functions. You could do this manually by finding the packages you need one by one or get a distribution that installs everything you need (including the programming language) like Anaconda for Python.

Think of the packages as add-ons to the programming language—the programming language is the DVD player and now you need the actual DVDs! Anaconda has almost everything and comes with a nice IDLE that helps you write Python code. But for those that want to do this manually (which I think is better in the long run), the ones I most commonly use are:

Disclosure: The author works for H2O.ai.

• NumPy for statistics and data formats
• SciPy for optimization and sparse data formats/operations
• Pandas for data manipulation
• Matplotlib for graphs and plots
• Sklearn for machine learning in general
• H2O for both machine learning algorithms and data manipulation operations
• Keras with Theano or TensorFlow for deep learning/predictive modeling
• XGBoost and LightGBM for two of the most successful predictive algorithms for a vast variety of predictive problems out there.

The list may seem daunting for those starting from zero. But you don’t need to worry too much, as there is a big overlap between the packages. Once you learn one, you’ve also learned a bit from the others as well.

Learn the basics of linear algebra, statistical modeling, and machine learning

Linear algebra

There are many good resources for learning linear algebra. You can find them here, here, here, and here. The one I used when I started were the slides from UCL. However, I did have some knowledge of linear algebra prior to learning data science, so consider the other links too. You only need one, so just take your pick.

Hardware

Kaggle

Analytics Vidhya and other platforms

Diving deeper: Stacking

Stay updated