Book Review: A Hands-on Introduction to Machine Learning

I used to be happy to obtain a evaluation copy of this new title from Cambridge University Press, “A Hands-on Introduction to Machine Learning.” The hardcover e book could be very engaging, well-produced and stable! It will overwhelm your backpack for positive. As a college teacher myself, I instantly appreciated creator and University of Washington professor Chirag Shah‘s pedagogical strategy. This is a gainful studying instrument. Every chapter has wonderful protection of the everyday machine studying (ML) matters coupled with very useful “Try It Yourself” sidebars that permit readers to train their understanding of the themes as they progress by way of the fabric. Each chapter additionally features a “Conceptual Questions” and “Hands-on Examples” characteristic on the finish of every part. I additionally appreciated the “Further Reading and Resources” and “Notes” sources on the finish of every chapter. Most of the code lined within the e book makes use of the Python language though the code will not be supplied on the e book’s web site (a be aware to publishers: it needs to be a no brainer to supply downloadable code to save the reader’s time). There is a particular appendix “R for Machine Learning” that’s helpful to get one other perspective of coding for ML. All of this academic materials helps the reader deal with the e book as a self-contained ML course.

Great content material for these new to ML

For a normal look on ML, learners will respect Part II on Supervised Learning and Part III on Unsupervised Learning. Part II consists of matters: Chapter 4 on Regression, and Chapters 5 and 6 on Classification. Part III consists of matters: Chapter 7 on Clustering and Chapter 8 on Dimensionality Reduction. These chapters are foundational by way of directing the reader alongside the right path for fixing real-life issues utilizing the rules of machine studying. You’ll discover normal matters like accessing datasets, exploratory knowledge evaluation, visualization (with matplotlib), knowledge transformation and have engineering. The numpy and pandas libraries are used all through. Of course, you’ll use widespread ML algorithms from scikit-learn. All the code examples are easy and direct, precisely what a learner new to ML wants. The protection of common algorithms is nice, for instance the classification fashions included are: kNN, determination bushes, random forest, logistic regression, softmax regression, naive Bayes, and SVMs. On the unsupervised aspect, the e book consists of numerous clustering methods together with dimensionality discount (PCA and LDA).

Maybe my favourite space of the e book is Part IV on neural networks; partly it helps the reader achieve appreciation for what’s going with AI at this time – generative AI. Chapter 9 is a whirlwind introduction to NNs with brief dialogue and code examples for: perceptrons, multilayer perceptrons (MLPs), ahead/again propagation, autoencoders, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and lengthy short-term reminiscence (LSTM). The discussions of those necessary strategies solely open the door to these matters because the reader should interact sources outdoors of the e book to absolutely perceive how they work (a powerful dose of arithmetic included). Chapter 10 turns the nook towards deep studying and deep neural networks. You’ll see code examples of PyTorch utilizing the venerable MNIST dataset so as to construct a handwritten digits classifier. You’ll even be launched to embeddings together with encoders and transformers.

Part V of the e book is used to introduce the reader to quite a lot of up to date matters that you simply’ll hear talked about on a regular basis in 2024: reinforcement studying (RL) and accountable AI.

Instructor Materials

One doubtlessly huge characteristic of this e book is all the trainer supplies that the creator developed to be used alongside aspect the textual content in a category atmosphere. I see the e book as a superb selection for an introductory class in ML. The teacher supplies are mentioned to embrace: options to many issues and workouts within the textbook, an teacher handbook with educating strategies together with recommendation for educating utilizing this textbook in-person and on-line, pattern mid-term assessments, lecture slides for every chapter that instructors can use and adapt for educating, and figures/tables from the textbook. As keen as I used to be to consider these supplies, I used to be unable to entry them due to an excessively restrictive course of by the writer. Too unhealthy, as a result of I believe utilizing this e book as a textbook for a category in ML represents a big potential market that may thus go untapped.


All in all I think about this e book as a superb new entry into the sector of machine studying and deep studying. I plan to add this title to the bibliography I give to my starting knowledge science college students as an academic useful resource they will eat simply after taking my intro class. Kudos to creator Shah for seeing a necessity for the sort of textual content!

Contributed by Daniel D. Gutierrez, Editor-in-Chief and Resident Data Scientist for insideBIGDATA. In addition to being a tech journalist, Daniel is also a marketing consultant in knowledge scientist, creator, educator and sits on a lot of advisory boards for numerous start-up corporations. 

Sign up for the free insideBIGDATA e-newsletter.

Join us on Twitter:

Join us on LinkedIn:

Join us on Facebook:

Recommended For You