# Best Data Science Book 2021 + 3 Bonus Recommendations

You want to learn about data science algorithms but you hate reading boring theoretical books? Then I've got a few books for you which will have you coding alongside the content.

As a data scientist you need a broad understanding of the various ml algorithms out there and you also need a good understanding of how to create an ml project end-to-end. Hands-on machine learning with Scikit-Learn and TensorFlow is the perfect book to achieve both of those goals. What I love about the book is the fact that it's a code-long style of a book.

All the concepts in the book is taught through code. It really forces you to holistically engage with the book.

The best part is that there's a Github repo for this book that contains Jupyter notebooks that relate to each of the chapters of the book. This includes all of the examples that it used to explain the concepts in the book.

Another thing I love about the book is that it's completely applications driven. So for example if the author is talking about decision trees he'll go ahead and show you an example in code by using the Iris data set from sklearn. So he'll walk you through the problem of identifying the flower type based on the petal length and the petal width. This is an extremely engaging way of learning about decision trees.

At the end of each chapter, there's a quiz which is extremely helpful in reinforcing the topics that you've just learned. Let's do a review on what you'd learn in the book. So the book is split up into two parts.

## Part 1

The first part is the fundamentals of machine learning and the second part is more about neural networks and deep learning. The first section has eight chapters.

The first chapter is an introduction to machine learning. It'll cover all the basics of the different types of machine learning systems and the main challenges of ml like the data quality, under-fitting, overfitting and talk a little bit about testing and validation. This chapter is to give you an overview of the basics of ml. The second chapter is about the end-to-end requirements of a machine learning project. This chapter introduces you to different ways of framing the problem, cleaning your data, visualising your features, selecting the right algorithm, tuning your model system and finally monitoring it.

The third chapter is about classification. The fourth is about training model systems. This chapter talks about different types of regression algorithms, gradient descent, regularisation and learning curves. The next four chapters cover support vector machine, decision trees, ensemble learning and dimension reduction. Part one does a great job of giving you a good foundation in your data science journey.

## Part 2

Let's talk about part two. The first chapter will have you up and running with Tensorflow. The second will introduce you to the details of an artificial neural network. The third will be all about training deep neural networks and the nuances that come with it such as solving the issue of vanishing gradients, using pre-trained models, optimisers and techniques to avoid overfitting.

The third chapter is all about distributed training with Tensorflow. The next chapters are about convolutional neural networks, recurrent neural networks, auto encoders and finally reinforcement learning. Part one and two work really well to teach you everything from the basics to the advanced concepts of machine learning. It can be quite a ramp up especially if you're not exposed to these topics. I use this textbook often as a reference whenever I feel the need to refresh my understanding of ml concepts. So I couldn't recommend it highly enough.

## 3 Bonus Recommendations

Now let's talk about the three other books that I recommend that level up your data science journey significantly.

### Python Data Science Handbook

The first is the python data science handbook. This book will get you comfortable with using python for data science applications. Best part is that it's available for free online.

### Practical Statistics for Data Scientists

This book covers everything you need to know about statistics including exploratory data analysis, significance testing and sampling distributions.

### Build a career in data science

This book is very different to the technical books I've recommended. So far this book will help you in developing your soft skills that you need to excel as a data scientist and as a bonus it's also got a range of interview questions at the very end of the book. So be sure to check them out.