No. 137 of 500 No. 11 in Hands-On AI & Programming

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow

Concepts, Tools, and Techniques to Build Intelligent Systems

By Aurélien Géron

Print length
861 pages
Publisher
O'Reilly Media
Reading level
Intermediate
ISBN-13
978-1098125974
Published
November 8, 2022
Written for
Programmers comfortable in Python who want to build and train real machine learning models.

Synopsis

Aurélien Géron's book teaches machine learning through runnable code and worked projects rather than formal proofs. The first part works through the classical toolkit with Scikit-Learn: an end-to-end project on California housing data, classification, linear and polynomial regression, support vector machines, decision trees, random forests and other ensembles, dimensionality reduction, and unsupervised methods such as clustering and anomaly detection. The second part moves to neural networks with Keras and TensorFlow, covering training techniques for deep networks, custom models, data pipelines, convolutional networks for vision, recurrent networks and attention for sequences and language, autoencoders, generative adversarial networks, diffusion models, transformers, reinforcement learning, and deploying models at scale.

Each chapter pairs intuition and a modest amount of mathematics with Jupyter notebooks that readers can run, and ends with exercises; the companion repository for the third edition has drawn more than 14,000 stars on GitHub. The approach rewards readers who already program in Python and want to reach working models quickly, and the explanations of why techniques succeed or fail make it more than a cookbook. First published in 2017 and revised in 2019 and 2022, the book tracks the TensorFlow 2 ecosystem; Géron has since released a 2025 successor, Hands-On Machine Learning with Scikit-Learn and PyTorch, which readers starting today may prefer.

Scorecard

Editorial score

How it scored

Editorial score: 84.1 / 1005 weighted criteria

  • Influence30%

    27.0/30
  • Authority25%

    21.3/25
  • Practical value20%

    18.0/20
  • Currency15%

    9.8/15
  • Readability10%

    8.0/10
  • Editorial score

    84.1/100

How the rank was computed

Rank No. 137 of 500 comes from a composite of 60.6: observable signals 22.7 of 55 points plus editorial review 37.9 of 45 points.

All 11 inputs
InputPoints
Observable signals22.7/55
Reach (Amazon ratings volume)10.9/16.5
Consensus across best-of lists2.8/13.8
Awards and bestseller evidence0.0/8.3
Reader rating, sample-adjusted7.4/8.3
Amazon sales rank1.6/5.5
Recency0.0/2.8
Editorial review37.9/45
Author authority9.6/11.3
Influence12.2/13.5
Practical value8.1/9.0
Readability3.6/4.5
Currency4.4/6.8

Each criterion is scored from 1 to 10 and weighted, and the points add up to the editorial score. The rank combines observable signals (55 percent) with this score (45 percent). Read the methodology.

Verdict

A widely used, project-driven guide to machine learning in Python, though its TensorFlow focus now trails the field's move to PyTorch.

Strengths

  • Runnable notebooks and exercises for every chapter
  • Covers classical ML through transformers
  • Clear intuition without heavy mathematical proofs

Keep in mind

Its TensorFlow and Keras code is less aligned with today's PyTorch-dominated practice, and it gives limited coverage of large language model engineering.

Key takeaways

  1. The end-to-end project shows that framing the problem, exploring and cleaning data, and setting up sound validation matter as much as model choice.
  2. Classical methods such as random forests and gradient boosting remain strong baselines for tabular data and should be tried before reaching for deep learning.
  3. Training deep networks reliably depends on practical techniques such as careful initialization, batch normalization, better optimizers, learning-rate schedules and regularization like dropout.
  4. Modern architectures, from convolutional networks to transformers and diffusion models, can be understood and implemented incrementally with high-level Keras APIs.

Notable facts

About the author

Aurélien Géron

Aurélien Géron is a machine learning consultant and trainer who led YouTube's video classification team at Google from 2013 to 2016. He co-founded the French wireless internet provider Wifirst and served as its CTO from 2002 to 2012, and also founded the consulting firm Polyconseil. He has worked with clients in finance, defense and healthcare, and has lectured at Université Paris-Dauphine.

Questions

Is Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow still worth reading in 2026?
Yes for its first half, whose Scikit-Learn coverage of classical machine learning remains current and practical. The deep learning half uses TensorFlow and Keras, which remain in production use but have lost ground to PyTorch. Readers starting fresh may prefer Géron's 2025 edition built on Scikit-Learn and PyTorch, which keeps the same teaching approach.
What do I need to know before reading Hands-On Machine Learning?
You should be comfortable programming in Python. The book explains the mathematics it uses, and the companion repository includes refresher notebooks on NumPy, pandas, Matplotlib, linear algebra and calculus, so a high school math background is enough to start. Readers without programming experience will struggle, because nearly every concept is taught through code.
What does Hands-On Machine Learning cover?
The third edition spans the full practical curriculum: data preparation, regression, classification, support vector machines, decision trees, ensembles, dimensionality reduction and clustering with Scikit-Learn, then neural networks with Keras and TensorFlow, including convolutional and recurrent networks, attention and transformers, autoencoders, GANs, diffusion models, reinforcement learning, and training and deploying models at scale.

Sources

As an Amazon Associate, Best AI Books earns from qualifying purchases. Best AI Books is sponsored by Iternal.ai, whose founder and CEO, John Byron Hanby IV, wrote The AI Strategy Blueprint, our #1 ranked book. How we rank books.