In this interactive course, you’ll participate in my life stories, and learn data science tricks for optimizing your day-to-day life. You’ll make the perfect glass of lemonade using Thompson sampling. You’ll lose weight with differential equations. And you might just qualify for the Olympics with a bit of statistics!
In 2021, I wrote the book Everyday Data Science. This year, I’ve teamed up with Jim Fisher to transform the book into this course. Each chapter is an interactive tale, like a conversation with a storyteller. 📜 Here’s the first chapter to show you what we mean!
What’s in the course? 🎁
Ten interactive chapters. Each chapter is a self-contained case study, explaining a principle of data science. The first three chapters are available, and the first chapter is free! The next seven will drop over the coming months.
An Everyday Look At A/B Testing
Or, When Life Gives You Lemons 🍋
Everyone’s coming to your party this evening — but you’re out of lemonade! And what’s worse, you’ve forgotten the recipe! 😲 In this lesson, you’ll learn how to make rapid decisions using A/B testing. Along the way, you’ll learn how to get the most money from a multi-armed bandit, by modelling them with the Beta distribution and then using Thompson sampling.
(This lesson is free while stocks last!)
An Everyday Look At Differential Equations
Or, ODEs On A Diet! ⚖️
Last year, my bathroom scales put my BMI at : too low. In this lesson, you’ll design my new diet regimen by running ordinary differential equations in Python. Along the way, you’ll learn the predator-prey model of ecosystems, and the Forbes model of metabolism (you might be surprised how your own body works!).
An Everyday Look At Time Series
Or, The Way You Do That Walk 👟
After the party, your phone is missing! In this chapter, we find the thief just by using accelerometer data your phone is logging. Along the way, you’ll learn what a time series is, and how to describe one with the ARMA model. You will fit a model using Python and automatic minimization. And you’ll classify time series using their fitted parameters.
An Everyday Look at Imputation 🆕
Or: No, Garmin, I Didn't Die In My Sleep
Our smartwatches track all kinds of useful metrics. Except when they run out of battery, or we have to charge them. If you were to analyze my historical heart rate data, you might conclude that I die every five days. Not so!
In this chapter, you’ll learn ways to handle
NaN values in your dataset. You’ll learn how to impute them (that is, fill in the blanks) using a method called iterative imputation. But you’ll also learn the dangers of imputation, and some simple alternatives to imputation including listwise deletion and feature deletion.
An Everyday Look At Vectors
Or, Your Resumé Lives in
Imagine if you could write and get the result: . Math is not just memorization and algorithmic computation — it’s a study of relationships; a beautiful interplay between what we know, and what we hope is true. In this chapter, you’ll see the power of word vectors, and learn to use them to improve your resumé.
An Everyday Look At Populations
Or, Your Body 👨👨👧👦
How well do you fit in with the average? Average height, weight, intelligence? Averages are used to describe a population and are thereby applied frequently to individuals of that population. It can be dangerous to live by averages because what if you are an outlier? In health care especially, experts use well-tuned averages to make a diagnosis. But again, what if you are different?
In this chapter, we have a story of a man whose life was saved because he didn’t fit the average and spoke up.
An Everyday Look at Graphs
Or, Walking The Dog 🐕🦺
“Again? I just took you out!” This was a very common phrase with our senior cocker spaniel named Lady. We took her out so often, however, that we started to wonder if there was rhyme or reason to her asking.
My wife and I meticulously recorded her potty habits for two weeks to see if we could find anything interesting that might help us help her have a more enjoyable time.
An Everyday Look at Bayesian-Optimal Pricing 🆕
Or, Everything Must Go!
You're moving house and putting things on Facebook Marketplace. How do you sell things quickly, but still turn a good profit? That’s what Bayesian-optimal pricing is about.
An Everyday Look At Synthetic Data 🆕
Or, Grow Your Own
The word ‘data’ derives from the Latin for ‘something given to you’. But what if you don’t have friends to give you data for your birthday? Why, you can just create some new data yourself, of course!
In this chapter, we’ll use simulation to train a model without ever dirtying our hands in the real world. It’s just like the Construct in The Matrix.
An Everyday Look At Goals
Or, The Olympics is Calling 🏃♀️
Jared Ward’s thesis on optimal pacing strategies for the 2013 St George Marathon served as a guide to help him prepare and eventually qualify for the 2016 Rio Olympic games.
Jared rigorously recorded data about his training over the course of years. In this chapter, we look at how we can use data science to meet our fitness goals and how Jared used stats to qualify for the Olympics.
What people are saying …
⭐⭐⭐⭐⭐ “I just spent half an hour in an airport reading the first chapter. With only a bit of knowledge of Bernoulli distributions, I found it very informative and well written, moving forward at a nice pace. The tiny tests made me reread a few things I had glossed over initially, forcing me to understand it a little better. Good job.” – Learner on HN.
⭐⭐⭐⭐⭐ “I love the format. The interactive prompts are great. They really get you to engage with the content. The humorous tone strikes a good balance and doesn’t get distracting as well.” — Learner on HN.
⭐⭐⭐⭐⭐ “I find myself irritated by how effective this method of instruction is. Every university course should start with at least a day of instruction written in this style.” — Learner on HN.
The book was #1 on Hacker News, and the #1 best seller in CS and Tech Education books on Amazon. Here are some things people have said.
⭐⭐⭐⭐⭐ “Like Freakonomics, but Data Science. This is really fun! Lots of interesting case studies, got my brain running on personal data projects I could play around with.” – Taylor Sorensen on Amazon.
⭐⭐⭐⭐⭐ “Andrew is easily one of the most productive people I've worked with. His deep mathematics and machine learning knowledge make him incredibly valuable. Andrew led a machine learning business that I was a part of, and was instrumental in its success and acquisition.” – Josh Greaves, Google Brain.
⭐⭐⭐⭐⭐ “Perfect for Data Science Younglings. Had fantastic examples of how you can apply data science to problems you might encounter. As someone who loves to ‘tinker’, it provides a great launching point for trying out data science in your own life in a reasonable and very useful way. Would recommend 10/10.” – PH on Amazon.
⭐⭐⭐⭐⭐ “Andrew not only has brilliant ideas, but understands how to communicate those ideas to others in a way that gets them excited. Many of his ideas are brilliantly simple concepts that take a lot of talent and hard work to create.” – Andrew W. Daniels.
⭐ “There is a serious lack of any editing, to the point that certain chapters are a SINGLE WORD.” – Sangwhan Moon on Amazon, who apparently really hated that joke chapter.
⭐⭐⭐⭐⭐ “Andrew is the go-to guy for understanding the math and statistics behind state of the art machine learning.” – Joshua Mathias.
⭐⭐⭐⭐⭐ “Andrew is a humble guy. He’s not the sort to add a wall of quotes to this page. I made him do it.” – Jim Fisher, TigYog.
👋 Hi, I’m Andrew. I’ll be your guide.
I’m a computational mathematician that specializes in machine learning. I wrote code for self-driving cars, and developed patentable natural language technology. Here’s my website and here’s my Twitter. Follow me for data science memes.
I was advised by David Wingate during my undergrad and master's programs, after which I joined OpenAI as a fellow. I spent some time as a research intern at Google Brain working with Quentin Berthet on applications of differentiable programming to self supervised learning on audio. I'm broadly interested in optimal transport, program synthesis, generative modeling, and theory of machine learning. I currently work on generative modeling for privacy and program synthesis!
For your money, you’ll get access to:
⭐ All seven original chapters, completely reworked as interactive lessons. These will be published over the next few months.
⭐ Three completely new interactive lessons!
⭐ Full source code access: Colab notebooks for all Python exercises
⭐ Certificate of completion 🎓
⭐ 30-day money-back guarantee
Frequently Asked Questions ()
“What are the prerequisites?” I try to make the course as approachable as possible. You should know what functions are in both math and programming. You should have some basic knowledge of stats, and a bit of experience with calculus concepts like derivatives (although I do some refreshers along the way).
But the most important prerequisite is a desire to work with data. Everything else you can figure out as you go. And I'm always open to 1-1 emails and chats if you get stuck or want to go deeper! 🙂
“What is Data Science?” Data Science is making decisions with data. It's not a magic cure-all, but it might help you optimize your life. If you take this course, you’ll see many fun case studies where you can use data in everyday situations, with some interesting math and tricks.
“How is Data Science?” Don’t you mean, “How is Data Science done?” Well, you’ll find out in this course by doing some! But here’s an example. Why hasn't there been a new Pirates of the Caribbean movie recently? 🏴☠️ I had a theory: you only make a sequel if the last movie was well-received. I plotted their Rotten Tomatoes scores:
Imagine you’re in charge of giving the green light for another sequel. Now you’ve seen this graph, what would you do? If you said “reject the sequel”, you might want to do some more data science, because the profit per film tells a slightly different story! 💸 We’ll see in this course that asking the right question is critical!
“Why is Data Science?” More than half of all organizations suffer from data illiteracy. People in the organization don’t understand what the data means, or how to use it to make decisions. And the decision makers – managers and company leaders – are even less likely to be data literate!
“Who is Data Science?” You! This course is for you!
“When is Data Science?” Every day
“What if I don’t like the course?” Then … you don’t have to buy it. But if you buy it and then you don’t like it, you can get a refund, no questions asked 🙂
“What the frick is TigYog?” TigYog lets anyone write interactive courses just like this one! Everyday Data Science is the first flagship course on TigYog. But you could write the next one! Learn more.
“You didn’t cover my favorite algorithm!” That’s not a question, and this course is not a textbook. 📚 It’s a collection of interesting tales. 📜 But they might inspire you to pick up a textbook afterwards – reach out and I'll give you my top recommendations!
“Will content be updated continuously?” I will be publishing ten chapters eventually. The topics are mostly timeless, so I don’t plan to update it after that. The jokes and cultural references will gracefully go stale.
“I’m stuck; where can I go for help?” Don’t feel bad about making a guess if you have to! If something is still unclear, reach out to me on Twitter, and I’ll try to make it right.