Python Deep Learning: Understand how deep neural networks work and apply them to real-world tasks

Cover
Title Page
Copyright and Credit
Contributors
Table of Contents
Preface
Part 1: Introduction to Neural Networks
Chapter 1: Machine Learning – an Introduction
	Technical requirements
	Introduction to ML
	Different ML approaches
		Supervised learning
		Unsupervised learning
		Reinforcement learning
		Components of an ML solution
		Neural networks
		Introducing PyTorch
	Summary
Chapter 2: Neural Networks
	Technical requirements
	The need for NNs
	The math of NNs
		Linear algebra
		An introduction to probability
		Differential calculus
	An introduction to NNs
		Units – the smallest NN building block
		Layers as operations
		Multi-layer NNs
		Activation functions
		The universal approximation theorem
	Training NNs
		GD
		Backpropagation
		A code example of an NN for the XOR function
	Summary
Chapter 3: Deep Learning Fundamentals
	Technical requirements
	Introduction to DL
	Fundamental DL concepts
		Feature learning
		The reasons for DL’s popularity
	Deep neural networks
	Training deep neural networks
		Improved activation functions
		DNN regularization
	Applications of DL
	Introducing popular DL libraries
		Classifying digits with Keras
		Classifying digits with PyTorch
	Summary
Part 2: Deep Neural Networks for Computer Vision
Chapter 4: Computer Vision with Convolutional Networks
	Technical requirements
	Intuition and justification for CNNs
	Convolutional layers
		A coding example of the convolution operation
		Cross-channel and depthwise convolutions
		Stride and padding in convolutional layers
	Pooling layers
	The structure of a convolutional network
	Classifying images with PyTorch and Keras
		Convolutional layers in deep learning libraries
		Data augmentation
		Classifying images with PyTorch
		Classifying images with Keras
	Advanced types of convolutions
		1D, 2D, and 3D convolutions
		1×1 convolutions
		Depthwise separable convolutions
		Dilated convolutions
		Transposed convolutions
	Advanced CNN models
		Introducing residual networks
		Inception networks
		Introducing Xception
		Squeeze-and-Excitation Networks
		Introducing MobileNet
		EfficientNet
		Using pre-trained models with PyTorch and Keras
	Summary
Chapter 5: Advanced Computer Vision Applications
	Technical requirements
	Transfer learning (TL)
		Transfer learning with PyTorch
		Transfer learning with Keras
	Object detection
		Approaches to object detection
		Object detection with YOLO
		Object detection with Faster R-CNN
	Introducing image segmentation
		Semantic segmentation with U-Net
		Instance segmentation with Mask R-CNN
	Image generation with diffusion models
		Introducing generative models
		Denoising Diffusion Probabilistic Models
	Summary
Part 3: Natural Language Processing and Transformers
Chapter 6: Natural Language Processing and Recurrent Neural Networks
	Technical requirements
	Natural language processing
		Tokenization
		Introducing word embeddings
		Word2Vec
		Visualizing embedding vectors
		Language modeling
	Introducing RNNs
		RNN implementation and training
		Backpropagation through time
		Vanishing and exploding gradients
		Long-short term memory
		Gated recurrent units
	Implementing text classification
	Summary
Chapter 7: The Attention Mechanism and Transformers
	Technical requirements
	Introducing seq2seq models
	Understanding the attention mechanism
		Bahdanau attention
		Luong attention
		General attention
		Transformer attention
		Implementing TA
	Building transformers with attention
		Transformer encoder
		Transformer decoder
		Putting it all together
		Decoder-only and encoder-only models
		Bidirectional Encoder Representations from Transformers
		Generative Pre-trained Transformer
	Summary
Chapter 8: Exploring Large Language Models in Depth
	Technical requirements
	Introducing LLMs
	LLM architecture
		LLM attention variants
		Prefix decoder
		Transformer nuts and bolts
		Models
	Training LLMs
		Training datasets
		Pre-training properties
		FT with RLHF
	Emergent abilities of LLMs
	Introducing Hugging Face Transformers
	Summary
Chapter 9: Advanced Applications of Large Language Models
	Technical requirements
	Classifying images with Vision Transformer
		Using ViT with Hugging Face Transformers
	Understanding the DEtection TRansformer
		Using DetR with Hugging Face Transformers
	Generating images with stable diffusion
		Autoencoder
		Conditioning transformer
		Diffusion model
		Using stable diffusion with Hugging Face Transformers
	Exploring fine-tuning transformers
	Harnessing the power of LLMs with LangChain
		Using LangChain in practice
	Summary
Part 4: Developing and Deploying Deep Neural Networks
Chapter 10: Machine Learning Operations (MLOps)
	Technical requirements
	Understanding model development
		Choosing an NN framework
		PyTorch versus TensorFlow versus JAX
		Open Neural Network Exchange
		Introducing TensorBoard
		Developing NN models for edge devices with TF Lite
		Mixed-precision training with PyTorch
	Exploring model deployment
		Deploying NN models with Flask
		Building ML web apps with Gradio
	Summary
Index
Other Books You May Enjoy