New MEAP! Save half on Deep Learning with Python, Third Edition

7 chapters of this new MEAP are available now, with more to follow soon!

Chapter 1: Start with the big question—what is deep learning? Journey through the history of machine learning and AI, and discover why deep learning has become so pivotal in recent years.

Chapter 2: Unpack the mathematical foundations of neural networks. Get hands-on with your first neural network, learn how data is represented, and explore the core operations that power these networks.

Chapter 3: Set up your deep learning workspace and take your first steps with TensorFlow, JAX, and PyTorch. Explore the anatomy of a neural network using core Keras APIs, and get to grips with MLOps.

Chapter 4: Apply your knowledge to real-world problems. Tackle binary classification by analyzing movie reviews, venture into multiclass classification with newswire categorization, and predict house prices in a regression example.

Engaging diagrams and figures make complex concepts easy to understand.

Chapter 5: Master the fundamentals of machine learning. Understand the importance of generalization, learn to evaluate models effectively, and discover techniques to improve both model fit and generalization.

Chapter 6: Learn the universal workflow of machine learning. From defining tasks to developing and deploying models, gain a comprehensive understanding of the entire process.

Chapter 7: Take a deep dive into Keras. Engage with various workflows, discover different ways to build models, and master both built-in and custom training and evaluation loops.

Chapter 8: Unlock the power of deep learning for computer vision. Start with convnets and image classification, then progress to advanced topics like image segmentation and object detection. Learn to leverage pretrained models like Segment Anything and YOLOv8.

Chapter 9: Harness deep learning for time series analysis. Explore various time series tasks, work through a temperature-forecasting example, and gain advanced knowledge of recurrent neural networks.

Chapter 10: From basic text preparation to advanced transformer architectures, master the techniques driving modern language models. Build your own GPT, understand the power of attention, and explore the fascinating world of generative AI.

Chapter 11: Best practices that ensure your models perform in the real world. Learn to optimize your models through hyperparameter tuning and ensembling. Scale up your training with distributed techniques and specialized hardware. Dive into advanced topics like model parallelism, LORA, and quantization for efficient large language model deployment.

François Chollet is a software engineer at Google, the creator of the Keras deep learning library, and has been named by TIME as one of the 100 most influential people in AI. In 2024, Chollet launched ARC Prize, a $1 million competition to solve the ARC-AGI benchmark.

Matthew Watson has been working on machine learning systems across Google since 2018.