Book Image

The Deep Learning Architect's Handbook

By : Ee Kin Chin
5 (1)
Book Image

The Deep Learning Architect's Handbook

5 (1)
By: Ee Kin Chin

Overview of this book

Deep learning enables previously unattainable feats in automation, but extracting real-world business value from it is a daunting task. This book will teach you how to build complex deep learning models and gain intuition for structuring your data to accomplish your deep learning objectives. This deep learning book explores every aspect of the deep learning life cycle, from planning and data preparation to model deployment and governance, using real-world scenarios that will take you through creating, deploying, and managing advanced solutions. You’ll also learn how to work with image, audio, text, and video data using deep learning architectures, as well as optimize and evaluate your deep learning models objectively to address issues such as bias, fairness, adversarial attacks, and model transparency. As you progress, you’ll harness the power of AI platforms to streamline the deep learning life cycle and leverage Python libraries and frameworks such as PyTorch, ONNX, Catalyst, MLFlow, Captum, Nvidia Triton, Prometheus, and Grafana to execute efficient deep learning architectures, optimize model performance, and streamline the deployment processes. You’ll also discover the transformative potential of large language models (LLMs) for a wide array of applications. By the end of this book, you'll have mastered deep learning techniques to unlock its full potential for your endeavors.

Related Content you might be interested in

Current Title:

Small book image
The Deep Learning Architect's Handbook
Ee Kin Chin
Dec, 2023 | 516 pages
Small book image
Responsible AI in the Enterprise
Adnan Masood | Heather Dawe
Jul, 2023 | 318 pages
Small book image
Generative AI with LangChain
Ben Auffarth
Dec, 2023 | 360 pages
Small book image
Deep Learning for Genomics
Upendra Kumar Devisetty
Nov, 2022 | 270 pages
Table of Contents (25 chapters)
Preface
Preface
Who this book is for
What this book covers
To get the most out of this book
Download the example code files
Conventions used
Get in touch
Share Your Thoughts
Share Your Thoughts
Download a free PDF copy of this book
1
Part 1 – Foundational Methods
Part 1 – Foundational Methods
Free Chapter
2
Chapter 1: Deep Learning Life Cycle
Chapter 1: Deep Learning Life Cycle
Technical requirements
Understanding the machine learning life cycle
Strategizing the construction of a deep learning system
Preparing data
Developing deep learning models
Delivering model insights
Managing risks
Summary
Further reading
3
Chapter 2: Designing Deep Learning Architectures
Chapter 2: Designing Deep Learning Architectures
Technical requirements
Exploring the foundations of neural networks using an MLP
Understanding neural network gradients
Understanding gradient descent
Implementing an MLP from scratch
Summary
4
Chapter 3: Understanding Convolutional Neural Networks
Chapter 3: Understanding Convolutional Neural Networks
Technical requirements
Understanding the convolutional neural network layer
Understanding the pooling layer
Building a CNN architecture
Designing a CNN architecture for practical usage
Exploring the CNN architecture families
Summary
5
Chapter 4: Understanding Recurrent Neural Networks
Chapter 4: Understanding Recurrent Neural Networks
Technical requirements
Understanding LSTM
Understanding GRU
Understanding advancements over the standard GRU and LSTM layers
Summary
6
Chapter 5: Understanding Autoencoders
Chapter 5: Understanding Autoencoders
Technical requirements
Decoding the standard autoencoder
Exploring autoencoder variations
Building a CNN autoencoder
Summary
7
Chapter 6: Understanding Neural Network Transformers
Chapter 6: Understanding Neural Network Transformers
Exploring neural network transformers
Decoding the original transformer architecture holistically
Uncovering transformer improvements using only the encoder
Uncovering transformer improvements using only the decoder
Summary
8
Chapter 7: Deep Neural Architecture Search
Chapter 7: Deep Neural Architecture Search
Technical requirements
Understanding the big picture of NAS
Understanding general hyperparameter search-based NAS
Understanding RL-based NAS
Understanding non-RL-based NAS
Summary
9
Chapter 8: Exploring Supervised Deep Learning
Chapter 8: Exploring Supervised Deep Learning
Technical requirements
Exploring supervised use cases and problem types
Implementing neural network layers for foundational problem types
Training supervised deep learning models effectively
Exploring general techniques to realize and improve supervised deep learning based solutions
Breaking down the multitask paradigm in supervised deep learning
Summary
10
Chapter 9: Exploring Unsupervised Deep Learning
Chapter 9: Exploring Unsupervised Deep Learning
Technical requirements
Exploring unsupervised deep learning applications
Creating pretrained network weights for downstream tasks
Creating general representations through unsupervised deep learning
Exploring zero-shot learning
Exploring the dimensionality reduction component of unsupervised deep learning
Detecting anomalies in external data
Summary
11
Part 2 – Multimodal Model Insights
Part 2 – Multimodal Model Insights
12
Chapter 10: Exploring Model Evaluation Methods
Chapter 10: Exploring Model Evaluation Methods
Technical requirements
Exploring the different model evaluation methods
Engineering the base model evaluation metric
Exploring custom metrics and their applications
Exploring statistical tests for comparing model metrics
Relating the evaluation metric to success
Directly optimizing the metric
Summary
13
Chapter 11: Explaining Neural Network Predictions
Chapter 11: Explaining Neural Network Predictions
Technical requirements
Exploring the value of prediction explanations
Demystifying prediction explanation techniques
Exploring gradient-based prediction explanations
Trusting and understanding integrated gradients
Using integrated gradients to aid in understanding predictions
Explaining prediction explanations automatically
Exploring common pitfalls in prediction explanations and how to avoid them
Summary
Further reading
14
Chapter 12: Interpreting Neural Networks
Chapter 12: Interpreting Neural Networks
Technical requirements
Interpreting neurons
Finding neurons to interpret
Interpreting learned image patterns
Discovering the counterfactual explanation strategy
Summary
15
Chapter 13: Exploring Bias and Fairness
Chapter 13: Exploring Bias and Fairness
Technical requirements
Exploring the types of bias
Understanding the source of AI bias
Discovering bias and fairness evaluation methods
Evaluating the bias and fairness of a deep learning model
Tailoring bias and fairness measures across use cases
Mitigating AI bias
Summary
16
Chapter 14: Analyzing Adversarial Performance
Chapter 14: Analyzing Adversarial Performance
Technical requirements
Using data augmentations for adversarial analysis
Analyzing adversarial performance for audio-based models
Analyzing adversarial performance for image-based models
Exploring adversarial analysis for text-based models
Summary
17
Part 3 – DLOps
Part 3 – DLOps
18
Chapter 15: Deploying Deep Learning Models to Production
Chapter 15: Deploying Deep Learning Models to Production
Technical requirements
Exploring the crucial components for DL model deployment
Identifying key DL model deployment requirements
Choosing the right DL model deployment options
Exploring deployment decisions based on practical use cases
Discovering general recommendations for DL deployment
Deploying a language model with ONNX, TensorRT, and NVIDIA Triton Server
Summary
19
Chapter 16: Governing Deep Learning Models
Chapter 16: Governing Deep Learning Models
Technical requirements
Governing deep learning model utilization
Governing a deep learning model through monitoring
Governing a deep learning model through maintenance
Summary
20
Chapter 17: Managing Drift Effectively in a Dynamic Environment
Chapter 17: Managing Drift Effectively in a Dynamic Environment
Technical requirements
Exploring the issues of drift
Exploring the types of drift
Exploring strategies to handle drift
Detecting drift programmatically
Summary
21
Chapter 18: Exploring the DataRobot AI Platform
Chapter 18: Exploring the DataRobot AI Platform
Technical requirements
A high-level look into what the DataRobot AI platform provides
Preparing data with DataRobot
Executing modeling experiments with DataRobot
Deploying a deep learning blueprint
Governing a deployed deep learning blueprint
Exploring some customer success stories
Summary
22
Chapter 19: Architecting LLM Solutions
Chapter 19: Architecting LLM Solutions
Overview of LLM solutions
Handling knowledge for LLM solutions
Evaluating LLM solutions
Identifying challenges with LLM solutions
Tackling challenges with LLM solutions
Leveraging LLM to build autonomous agents
Exploring LLM solution use cases
Summary
Further reading
23
Index
Index
Why subscribe?
24
Other Books You May Enjoy
Other Books You May Enjoy
Packt is searching for authors like you
Share Your Thoughts
Download a free PDF copy of this book

Preface

As a deep learning practitioner and enthusiast, I have spent years working on various projects and learning from diverse sources such as Kaggle, GitHub, colleagues, and real-life use cases. I've realized that there is a significant gap in the availability of cohesive, end-to-end deep learning resources. Traditional Massively Open Online Courses (MOOC), while helpful, often lack the practical knowledge and real-world insights that can only be gained through hands-on experience.

To bridge this gap, I've created The Deep Learning Architect Handbook, a comprehensive and practical guide that combines my unique experiences and insights. This book will help you navigate the complex landscape of deep learning, providing you with the knowledge and insights that would typically take years of hands-on experience to acquire, condensed into a resource that can be consumed in just days or weeks.

This book delves into various stages of the deep learning life cycle, from planning and data preparation to model deployment and governance. Throughout this journey, you'll encounter both foundational and advanced deep learning architectures, such as Multi-Layer Perceptrons (MLPs), Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), autoencoders, transformers, and cutting-edge methods, such as Neural Architecture Search (NAS). Divided into three parts, this book covers foundational methods, model insights, and DLOps, exploring advanced topics such as NAS, adversarial performance, and Large Language Model (LLM) solutions. By the end of this book, you will be well-prepared to design, develop, and deploy effective deep learning solutions, unlocking their full potential and driving innovation across various applications.

I hope that this book will serve as a way for me to give back to the community, by sparking conversations, challenging assumptions, and inspiring new ideas and approaches in the field of deep learning. I invite you to join me on this journey, and I look forward to hearing your thoughts and feedback as we explore the captivating world of deep learning together. Please feel free to reach out to me via LinkedIn through www.linkedin.com/in/chineekin, Kaggle through https://www.kaggle.com/dicksonchin93, or other channels listed on my LinkedIn profile. Your unique experiences and perspectives will undoubtedly contribute to the ongoing evolution of this book and the deep learning community as a whole.

Previous Section
Next Section