Packt Publishing has turned another one of my video courses, Training Your Systems with Python Statistical Modeling, into a book! This book is now available for purchase.
Training Systems using Python Statistical Modeling is now available from Packt Publishing’s website and from Amazon. This book was created by a team at Packt Publishing who took my video course and turned it into book form. If you’re like me and love books that you can hold in your hand, touch, thumb through, etc., and you’re looking to learn about statistics and machine learning methodology as used in Python, give my book a look.
My previous book, Hands-On Data Analysis with NumPy and Pandas, covered the basics of managing data sets in Python using two common tools, NumPy and pandas, along with how to use common tools for Python data analysis such as Anaconda and Jupyter Notebooks. Training Systems using Python Statistical Modeling follows naturally from that book, going from just managing data in Python to drawing inferences and developing useful applications from that data.
In addition to NumPy and pandas, this book shows how to use Scikit-Learn, SciPy, and statsmodels for statistical inference and machine learning tasks in Python. I start the book discussing basic statistical inference, including hypothesis testing and parameter intervals. This section uses statsmodels the most. Then the book progresses to supervised learning. I start with explaining basic concepts of supervised learning and how one should choose, tune, and evaluate supervised learning algorithms. After these basics I present learning models for classification and regression using scikit-learn, including decision trees, support vector machines, logistic regression, linear regression, Ridge/LASSO regression, and neural networks. After supervised learning comes unsupervised learning. This includes clustering methods such as the k-means algorithm and dimensionality reduction techniques such as principle component analysis. And this is only a broad overview of the techniques and algorithms the book covers; there are far more than I’ve mentioned here!
While the book offers basic explanations about how these methods work, it is not a theoretical book; it primarily moves forward with real data examples. Readers not only see a description of the methods and basic parameters but how to apply them to real data sets. In many ways this is a hands-on introduction to machine learning and statistics. Thus it’s a great resource for anyone looking to start right away developing data models in Python.
I would like to add that of the books I worked on I found this book the most enjoyable to write. While Hands-On Data Analysis with NumPy and Pandas is my personal best-seller it was frankly boring to write; its content is extremely basic and, while essential, does not show the powerful things people can do with data. This book, in comparison, finally shows how data can be useful and shows how to do interesting things with it. It also covers much more material, being nearly twice as long as Hands-On Data Analysis with NumPy and Pandas.
I list the book’s chapters below:
- Classical Statistical Analysis
- Introduction to Supervised Learning
- Binary Prediction Models
- Regression Analysis and How to Use It
- Neural Networks
- Clustering Techniques
- Dimensionality Reduction
Also, check out the book’s GitHub page to see code samples used in the book.
I would like to thank the staff at Packt Publishing for their work on this book, particularly Joseph Sunil. I was so pleased when I received my copies in the mail and I thank them for their hard work to make this possible.
The MSRP for the book is $27.99, but is currenlty on sale for $19.59 (30% off) on Packt’s website as part of their summer sale, so pick it up while it’s cheap! If you’re not interested in buying this particular book but still want access to it, perhaps consider getting a Mapt subscription. You’ll have access to thousands of books and video courses (including all of my content), and can even get one book to keep for free (without DRM) every month! Perhaps that book will be mine! It’s a great deal you should consider. Also, if you read the book, please leave a review (good or bad). These reviews help others decide whether this book is right for them. Good reviews help earn me sales. “Dissatisfied” reviews give me and Packt’s team feedback for later work (or perhaps to errata the book; there have been mistakes in it that slipped through the editing process).