Deep Learning with Python

Deep Learning with Python
brief contents
contents
preface
acknowledgments
about this book
Who should read this book
About the code
liveBook discussion forum
about the author
about the cover illustration
1 What is deep learning?
1.1 Artificial intelligence, machine learning, and deep learning
1.1.1 Artificial intelligence
1.1.2 Machine learning
1.1.3 Learning rules and representations from data
1.1.4 The “deep” in “deep learning”
1.1.5 Understanding how deep learning works, in three figures
1.1.6 What deep learning has achieved so far
1.1.7 Don’t believe the short-term hype
1.1.8 The promise of AI
1.2 Before deep learning: A brief history of machine learning
1.2.1 Probabilistic modeling
1.2.2 Early neural networks
1.2.3 Kernel methods
1.2.4 Decision trees, random forests, and gradient boosting machines
1.2.5 Back to neural networks
1.2.6 What makes deep learning different
1.2.7 The modern machine learning landscape
1.3 Why deep learning? Why now?
1.3.1 Hardware
1.3.2 Data
1.3.3 Algorithms
1.3.4 A new wave of investment
1.3.5 The democratization of deep learning
1.3.6 Will it last?
2 The mathematical building blocks of neural networks
2.1 A first look at a neural network
2.2 Data representations for neural networks
2.2.1 Scalars (rank-0 tensors)
2.2.2 Vectors (rank-1 tensors)
2.2.3 Matrices (rank-2 tensors)
2.2.4 Rank-3 and higher-rank tensors
2.2.5 Key attributes
2.2.6 Manipulating tensors in NumPy
2.2.7 The notion of data batches
2.2.8 Real-world examples of data tensors
2.2.9 Vector data
2.2.10 Timeseries data or sequence data
2.2.11 Image data
2.2.12 Video data
2.3 The gears of neural networks: Tensor operations
2.3.1 Element-wise operations
2.3.2 Broadcasting
2.3.3 Tensor product
2.3.4 Tensor reshaping
2.3.5 Geometric interpretation of tensor operations
2.3.6 A geometric interpretation of deep learning
2.4 The engine of neural networks: Gradient-based optimization
2.4.1 What’s a derivative?
2.4.2 Derivative of a tensor operation: The gradient
2.4.3 Stochastic gradient descent
2.4.4 Chaining derivatives: The Backpropagation algorithm
2.5 Looking back at our first example
2.5.1 Reimplementing our first example from scratch in TensorFlow
2.5.2 Running one training step
2.5.3 The full training loop
2.5.4 Evaluating the model
Summary
3 Introduction to Keras and TensorFlow
3.1 What’s TensorFlow?
3.2 What’s Keras?
3.3 Keras and TensorFlow: A brief history
3.4 Setting up a deep learning workspace
3.4.1 Jupyter notebooks: The preferred way to run deep learning experiments
3.4.2 Using Colaboratory
3.5 First steps with TensorFlow
3.5.1 Constant tensors and variables
3.5.2 Tensor operations: Doing math in TensorFlow
3.5.3 A second look at the GradientTape API
3.5.4 An end-to-end example: A linear classifier in pure TensorFlow
3.6 Anatomy of a neural network: Understanding core Keras APIs
3.6.1 Layers: The building blocks of deep learning
3.6.2 From layers to models
3.6.3 The “compile” step: Configuring the learning process
3.6.4 Picking a loss function
3.6.5 Understanding the fit() method
3.6.6 Monitoring loss and metrics on validation data
3.6.7 Inference: Using a model after training
Summary
4 Getting started with neural networks: Classification and regression
4.1 Classifying movie reviews: A binary classification example
4.1.1 The IMDB dataset
4.1.2 Preparing the data
4.1.3 Building your model
4.1.4 Validating your approach
4.1.5 Using a trained model to generate predictions on new data
4.1.6 Further experiments
4.1.7 Wrapping up
4.2 Classifying newswires: A multiclass classification example
4.2.1 The Reuters dataset
4.2.2 Preparing the data
4.2.3 Building your model
4.2.4 Validating your approach
4.2.5 Generating predictions on new data
4.2.6 A different way to handle the labels and the loss
4.2.7 The importance of having sufficiently large intermediate layers
4.2.8 Further experiments
4.2.9 Wrapping up
4.3 Predicting house prices: A regression example
4.3.1 The Boston housing price dataset
4.3.2 Preparing the data
4.3.3 Building your model
4.3.4 Validating your approach using K-fold validation
4.3.5 Generating predictions on new data
4.3.6 Wrapping up
Summary
5 Fundamentals of machine learning
5.1 Generalization: The goal of machine learning
5.1.1 Underfitting and overfitting
5.1.2 The nature of generalization in deep learning
5.2 Evaluating machine learning models
5.2.1 Training, validation, and test sets
5.2.2 Beating a common-sense baseline
5.2.3 Things to keep in mind about model evaluation
5.3 Improving model fit
5.3.1 Tuning key gradient descent parameters
5.3.2 Leveraging better architecture priors
5.3.3 Increasing model capacity
5.4 Improving generalization
5.4.1 Dataset curation
5.4.2 Feature engineering
5.4.3 Using early stopping
5.4.4 Regularizing your model
Summary
6 The universal workflow of machine learning
6.1 Define the task
6.1.1 Frame the problem
6.1.2 Collect a dataset
6.1.3 Understand your data
6.1.4 Choose a measure of success
6.2 Develop a model
6.2.1 Prepare the data
6.2.2 Choose an evaluation protocol
6.2.3 Beat a baseline
6.2.4 Scale up: Develop a model that overfits
6.2.5 Regularize and tune your model
6.3 Deploy the model
6.3.1 Explain your work to stakeholders and set expectations
6.3.2 Ship an inference model
6.3.3 Monitor your model in the wild
6.3.4 Maintain your model
Summary
7 Working with Keras: A deep dive
7.1 A spectrum of workflows
7.2 Different ways to build Keras models
7.2.1 The Sequential model
7.2.2 The Functional API
7.2.3 Subclassing the Model class
7.2.4 Mixing and matching different components
7.2.5 Remember: Use the right tool for the job
7.3 Using built-in training and evaluation loops
7.3.1 Writing your own metrics
7.3.2 Using callbacks
7.3.3 Writing your own callbacks
7.3.4 Monitoring and visualization with TensorBoard
7.4 Writing your own training and evaluation loops
7.4.1 Training versus inference
7.4.2 Low-level usage of metrics
7.4.3 A complete training and evaluation loop
7.4.4 Make it fast with tf.function
7.4.5 Leveraging fit() with a custom training loop
Summary
8 Introduction to deep learning for computer vision
8.1 Introduction to convnets
8.1.1 The convolution operation
8.1.2 The max-pooling operation
8.2 Training a convnet from scratch on a small dataset
8.2.1 The relevance of deep learning for small-data problems
8.2.2 Downloading the data
8.2.3 Building the model
8.2.4 Data preprocessing
8.2.5 Using data augmentation
8.3 Leveraging a pretrained model
8.3.1 Feature extraction with a pretrained model
8.3.2 Fine-tuning a pretrained model
Summary
9 Advanced deep learning for computer vision
9.1 Three essential computer vision tasks
9.2 An image segmentation example
9.3 Modern convnet architecture patterns
9.3.1 Modularity, hierarchy, and reuse
9.3.2 Residual connections
9.3.3 Batch normalization
9.3.4 Depthwise separable convolutions
9.3.5 Putting it together: A mini Xception-like model
9.4 Interpreting what convnets learn
9.4.1 Visualizing intermediate activations
9.4.2 Visualizing convnet filters
9.4.3 Visualizing heatmaps of class activation
Summary
10 Deep learning for timeseries
10.1 Different kinds of timeseries tasks
10.2 A temperature-forecasting example
10.2.1 Preparing the data
10.2.2 A common-sense, non-machine learning baseline
10.2.3 Let’s try a basic machine learning model
10.2.4 Let’s try a 1D convolutional model
10.2.5 A first recurrent baseline
10.3 Understanding recurrent neural networks
10.3.1 A recurrent layer in Keras
10.4 Advanced use of recurrent neural networks
10.4.1 Using recurrent dropout to fight overfitting
10.4.2 Stacking recurrent layers
10.4.3 Using bidirectional RNNs
10.4.4 Going even further
Summary
11 Deep learning for text
11.1 Natural language processing: The bird’s eye view
11.2 Preparing text data
11.2.1 Text standardization
11.2.2 Text splitting (tokenization)
11.2.3 Vocabulary indexing
11.2.4 Using the TextVectorization layer
11.3 Two approaches for representing groups of words: Sets and sequences
11.3.1 Preparing the IMDB movie reviews data
11.3.2 Processing words as a set: The bag-of-words approach
11.3.3 Processing words as a sequence: The sequence model approach
11.4 The Transformer architecture
11.4.1 Understanding self-attention
11.4.2 Multi-head attention
11.4.3 The Transformer encoder
11.4.4 When to use sequence models over bag-of-words models
11.5 Beyond text classification: Sequence-to-sequence learning
11.5.1 A machine translation example
11.5.2 Sequence-to-sequence learning with RNNs
11.5.3 Sequence-to-sequence learning with Transformer
Summary
12 Generative deep learning
12.1 Text generation
12.1.1 A brief history of generative deep learning for sequence generation
12.1.2 How do you generate sequence data?
12.1.3 The importance of the sampling strategy
12.1.4 Implementing text generation with Keras
12.1.5 A text-generation callback with variable-temperature sampling
12.1.6 Wrapping up
12.2 DeepDream
12.2.1 Implementing DeepDream in Keras
12.2.2 Wrapping up
12.3 Neural style transfer
12.3.1 The content loss
12.3.2 The style loss
12.3.3 Neural style transfer in Keras
12.3.4 Wrapping up
12.4 Generating images with variational autoencoders
12.4.1 Sampling from latent spaces of images
12.4.2 Concept vectors for image editing
12.4.3 Variational autoencoders
12.4.4 Implementing a VAE with Keras
12.4.5 Wrapping up
12.5 Introduction to generative adversarial networks
12.5.1 A schematic GAN implementation
12.5.2 A bag of tricks
12.5.3 Getting our hands on the CelebA dataset
12.5.4 The discriminator
12.5.5 The generator
12.5.6 The adversarial network
12.5.7 Wrapping up
Summary
13 Best practices for the real world
13.1 Getting the most out of your models
13.1.1 Hyperparameter optimization
13.1.2 Model ensembling
13.2 Scaling-up model training
13.2.1 Speeding up training on GPU with mixed precision
13.2.2 Multi-GPU training
13.2.3 TPU training
Summary
14 Conclusions
14.1 Key concepts in review
14.1.1 Various approaches to AI
14.1.2 What makes deep learning special within the field of machine learning
14.1.3 How to think about deep learning
14.1.4 Key enabling technologies
14.1.5 The universal machine learning workflow
14.1.6 Key network architectures
14.1.7 The space of possibilities
14.2 The limitations of deep learning
14.2.1 The risk of anthropomorphizing machine learning models
14.2.2 Automatons vs. intelligent agents
14.2.3 Local generalization vs. extreme generalization
14.2.4 The purpose of intelligence
14.2.5 Climbing the spectrum of generalization
14.3 Setting the course toward greater generality in AI
14.3.1 On the importance of setting the right objective: The shortcut rule
14.3.2 A new target
14.4 Implementing intelligence: The missing ingredients
14.4.1 Intelligence as sensitivity to abstract analogies
14.4.2 The two poles of abstraction
14.4.3 The missing half of the picture
14.5 The future of deep learning
14.5.1 Models as programs
14.5.2 Blending together deep learning and program synthesis
14.5.3 Lifelong learning and modular subroutine reuse
14.5.4 The long-term vision
14.6 Staying up to date in a fast-moving field
14.6.1 Practice on real-world problems using Kaggle
14.6.2 Read about the latest developments on arXiv
14.6.3 Explore the Keras ecosystem
Final words
index
Symbols
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X