liveBook · Manning

Model architecture

Also called neural network architecture, model’s architecture, deep learning architecture, neural architecture

Model architecture is the structural arrangement of a computational model’s components, connections, and operations that determines how it processes inputs and produces outputs.

How 22 Manning books explain it

Grokking Machine Learning

The structural specification of a neural network, including its layers, nodes, and activation functions.

How it is used
Uses connected classifiers to separate data that a single linear classifier cannot divide.
How deep it goes
Defines layer roles and connectivity, then explains how activation functions and successive classifiers shape decision boundaries.
Examples
Uses alien word counts and connected career, family, and happiness classifiers to illustrate network structure.
What you can do after
Understand how connected perceptrons and layers combine to classify data with more complex boundaries.

Read how Grokking Machine Learning explains model architecture →

Delivering Machine Learning Projects: From design to deployment

A model architecture is the structural arrangement of components and connections that defines how a machine learning model processes inputs and produces outputs.

How it is used
Places the model within a production system whose infrastructure constrains delivery.
How deep it goes
Describes high-level system components and feasibility checks but does not specify model internals.
Examples
Uses a smart building system with sensors, a database, model execution, and actuators to illustrate deployment components.
What you can do after
Evaluate a high-level deployment design for missing components or skills before implementation.

Read how Delivering Machine Learning Projects: From design to deployment explains model architecture →

Math and Architectures of Deep Learning

The chosen arrangement of neurons and their connections in a neural network.

How it is used
Frames model design as choosing a parameterized function before training estimates its weights.
How deep it goes
Defines linear architectures and relates neural-network neurons and connections to expressive power, while deferring popular architectures.
Examples
Uses a cat threat-score function with hardness, sharpness, two weights, and a bias.
What you can do after
Evaluate/understand how architecture choice differs from weight estimation and affects a network’s representational capacity.

Read how Math and Architectures of Deep Learning explains model architecture →

Lead Developer Career Guide

The high-level structural organization of a software system, including its components, interfaces, technologies, and interactions.

How it is used
Guides a lead developer’s technical approach and helps align project requirements, team estimates, budget, and system stability.
How deep it goes
Explains responsibilities, diagrams, and learning tactics, but does not specify component patterns or implementation details.
Examples
Uses a simple “Hello, World” application, a reusable prototype, and a lead developer–lead architect skills table.
What you can do after
Practice/perform architecture planning through a simple application and reusable prototype while seeking feedback from experienced developers.

Read how Lead Developer Career Guide explains model architecture →

Architecture Modernization: Socio-technical alignment of software, strategy, and structure

Software architecture is the fundamental structure of a software system, including its major components, relationships, and guiding design decisions.

How it is used
Connects software structure to business outcomes, team autonomy, and decisions about where to modernize.
How deep it goes
Works through domain decomposition and team alignment, but stops short of implementation-level design.
Examples
Uses an airline’s monolith and domain map to show how business goals can guide architectural boundaries.
What you can do after
Evaluate/understand it by comparing business domains, software boundaries, and team ownership to find change coupling.

Read how Architecture Modernization: Socio-technical alignment of software, strategy, and structure explains model architecture →

Build AI Drug Discovery Pipelines

A neural network architecture is the arrangement and connectivity of layers and neurons that defines a neural network's computational structure.

How it is used
Defines the network structure used to transform numerical inputs into predictions for tasks such as compound ranking.
How deep it goes
Explains layers, neurons, depth, width, dense connectivity, forward passes, and overfitting trade-offs, but stops short of advanced architecture design.
Examples
Uses layered-neuron diagrams, logistic regression, PyTorch tensors, and compound-ranking workflows as concrete examples.
What you can do after
Build it from scratch by specifying layers and connectivity in PyTorch, then fit, evaluate, modify, and save the model.

Read how Build AI Drug Discovery Pipelines explains model architecture →

Timeless Algorithms, The Foundational Ideas

An architecture is the structural organization of components and operations in a computational model or system.

How it is used
Frames architecture as the route through which a neural network transforms inputs into predictions.
How deep it goes
Explains layers, nodes, weights, biases and activations, then describes empirical comparison of architectural configurations.
Examples
Uses a two-hidden-node pedestrian-detection network and a separate braking threshold to distinguish prediction from action.
What you can do after
Evaluate how architectural choices affect capacity and compare candidate networks using validation and reserved test data.

Read how Timeless Algorithms, The Foundational Ideas explains model architecture →

Functional Design and Architecture: Examples in Haskell

Application architecture is the architecture of a single software application, including its major code structures and behavioral organization.

How it is used
Guides the organization and coordination of a single application's major components.
How deep it goes
Defines architecture qualities and component boundaries, then outlines iterative top-down design and diagramming without prescribing a mandatory notation.
Examples
Uses a spaceship control application to illustrate the integration risks of uncoordinated teams.
What you can do after
Evaluate and understand an application's architecture using qualities such as modularity, simplicity, robustness, and extensibility.

Read how Functional Design and Architecture: Examples in Haskell explains model architecture →

LLMs in Production

A model architecture is the structural design of a machine-learning model, including its computational components and their connections.

How it is used
Guides model definition before training, especially when building language models from scratch under task, data, and compute constraints.
How deep it goes
Defines layers and connections, demonstrates a small implementation, and stops short of deeply explaining modern architecture changes.
Examples
Uses a two-layer feed-forward network, a GPT-style module, and a small text-generation training run as concrete architecture examples.
What you can do after
Build it from scratch by selecting components, connecting them, defining loss, and integrating the result into a training loop.

Read how LLMs in Production explains model architecture →

Machine Learning Platform Engineering

A model architecture is the structural design of a machine learning model, including its computational components and their arrangement.

How it is used
Guides production model selection, training configuration, evaluation, and lifecycle tracking.
How deep it goes
Explains operational trade-offs and visualizes layers and connections, but stops short of deriving architecture internals.
Examples
Uses YOLOv8 object detection, pretrained weights, TensorBoard graphs, training inferences, and configurable training artifacts.
What you can do after
Evaluate and configure architectures through reproducible training runs, metric visualization, and artifact tracking.

Read how Machine Learning Platform Engineering explains model architecture →

Deep Learning with Python, Third Edition

The structural design of a deep learning model, here contrasted with approaches that adapt at test time rather than only scaling fixed architectures.

How it is used
Constrains the functions gradient descent can search by encoding assumptions about the problem.
How deep it goes
Defines layer choices and connections, explains their effect on learning, and leaves strong design choices largely to experience.
Examples
Uses linear separability, translation-invariant image patterns, and repeated Xception blocks as concrete architecture examples.
What you can do after
Build it from scratch by arranging layers and use Keras’s Functional API for graph-shaped models with multiple inputs or outputs.

Read how Deep Learning with Python, Third Edition explains model architecture →

Grokking AI Algorithms, Second Edition

A neural network architecture is the arrangement of computational layers, connections, inputs, and outputs that defines a neural network's structure.

How it is used
Guides the design and selection of neural networks for prediction, language modeling, and image generation.
How deep it goes
Explains configurable layers and outputs, then descends into Transformer dimensions and architecture tradeoffs without presenting implementation code.
Examples
Uses multilayer ANNs, multiple outputs, toy and real-world Transformer dimensions, and image-generation architecture requirements.
What you can do after
Evaluate architecture choices against a problem’s learning requirements, resource limits, latency, or visual-processing demands.

Read how Grokking AI Algorithms, Second Edition explains model architecture →

Fast Python: High performance techniques for large datasets

How it is used
Frames system architecture and implementation as places to seek performance gains when data growth makes processing more demanding.
How deep it goes
Identifies scaling and hardware constraints but stops short of detailed architecture designs or optimization procedures.
Examples
Contrasts scaling from one computer to multiple machines with moving an in-memory workload to disk.
What you can do after
Diagnose potential performance constraints by examining system architecture and implementation before adding machines.

Read how Fast Python: High performance techniques for large datasets explains model architecture →

Architecting for Autonomy

The architectural discipline that structures applications and exposes their capabilities and interfaces.

How it is used
Frames application architecture as the control point for actions and safe agent invocation.
How deep it goes
Explains interface contracts, authority boundaries, review, and recovery, without detailing implementation.
Examples
Uses a fraud agent freezing accounts, a failed payment recovery, and a regulated manufacturer’s lot-release decision.
What you can do after
Evaluate application interfaces and recovery paths for bounded autonomous actions.

Read how Architecting for Autonomy explains model architecture →

Vibe Engineering

A neural network architecture is the organization of computational layers and connections that transform inputs into outputs through learned parameters.

How it is used
Uses a sequential neural network to turn padded medical-text sequences into binary risk probabilities.
How deep it goes
Explains embedding, dropout, recurrent processing, and sigmoid output, including layer dimensions and dropout rate.
Examples
Illustrates drug-risk classification with a 20,000-word vocabulary, 180-token sequences, and a 64-dimensional embedding.
What you can do after
Understand how a sequence of neural network layers transforms text features into a risk probability.

Read how Vibe Engineering explains model architecture →

Coding with AI

Application architecture is the high-level organization of a software application's components, responsibilities, and interactions.

How it is used
Uses architecture planning to establish a foundation for application development and reduce maintenance costs.
How deep it goes
Develops from design planning into Flask project structure, separation of concerns, and human-led refinement.
Examples
Uses a certification practice-test design, a HAM radio exam app, and a suggested Flask file structure.
What you can do after
Evaluate and refine AI-generated structures and code against application needs, maintainability, and separation of concerns.

Read how Coding with AI explains model architecture →

The Art of Code

Software architecture is the high-level organization of a software system, including its components, relationships, boundaries, and governing decisions.

How it is used
Guides durability decisions by stabilizing exposed contracts and reviewing structural risks before changes affect dependent systems.
How deep it goes
Connects interface stability to codebase review signals, but leaves final architectural judgment dependent on broader system context.
Examples
Uses URL, header, or query-parameter versioning and flags large classes, circular dependencies, and frequently changed files.
What you can do after
Evaluate visible durability signals, gather missing context, and report architectural risks without treating automated review as final judgment.

Read how The Art of Code explains model architecture →

Machines that Think

A computational architecture is the structural organization of components and information flow that determines how a system processes inputs and produces outputs.

How it is used
Frames architecture as a hypothesis about information flow and a way to match neural computation to data structure.
How deep it goes
Contrasts major architectural families and their information flow, without implementation details.
Examples
Uses convolutional networks, recurrent networks, autoencoders, and generative models to illustrate different treatments of data.
What you can do after
Evaluate how architectural choices shape model capacity without assuming that scale guarantees performance.

Read how Machines that Think explains model architecture →

Grokking Machine Learning, Second Edition

The structural organization of layers, processing units, and connections that defines how a neural network transforms inputs into outputs.

How it is used
Defines how neural networks organize computation to fit data and solve problems.
How deep it goes
Explains layers, depth, bias nodes, connectivity, and classifier boundaries, but defers detailed alternative architectures.
Examples
Uses a 4-5-3-1 fully connected network, layered classifier boundaries, TensorFlow Playground, and a larger-network analogy.
What you can do after
Evaluate network diagrams by identifying layers, depth, node counts, connections, and architectural choices.

Read how Grokking Machine Learning, Second Edition explains model architecture →

Learn Generative AI with PyTorch

The internal design and structural organization of a neural model, here referring to the overall layout of an autoencoder.

How it is used
Frames generative models as inspectable systems whose design can guide better practical use.
How deep it goes
Explains architecture through construction goals and a GPT training-generation workflow, without detailing every internal component.
Examples
Uses from-scratch GAN and Transformer construction plus GPT prompt-based text generation as concrete architectural examples.
What you can do after
Build it from scratch by relating GPT architecture to token preparation, next-token training, and iterative text generation.

Read how Learn Generative AI with PyTorch explains model architecture →

The Complete Obsolete Guide to Generative AI

The structural design of a machine-learning model, including its components and their organization.

How it is used
Frames architecture as a design lever for tailoring language models to tasks, domains, privacy needs, and hardware constraints.
How deep it goes
Explains why architecture matters and how it relates to optimization, but stops short of describing specific components or construction steps.
Examples
Contrasts general-purpose GPT models with custom LLMs optimized for specialized tasks, domains, efficiency, or privacy.
What you can do after
Evaluate/understand why custom architecture choices can improve task fit, efficiency, privacy, and organizational control.

Read how The Complete Obsolete Guide to Generative AI explains model architecture →

Financial AI in Practice

The structural design of a predictive model, including its components and how they are organized.

How it is used
Frames financial AI as a designed system rather than a standalone model API call.
How deep it goes
Outlines multilayered workflows and design tradeoffs but leaves implementation details unspecified.
Examples
Contrasts credit-scoring trees, report-interpreting LLMs, and neural architectures for short-term trading.
What you can do after
Evaluate architecture choices against reasoning quality, operating cost, and governance needs.

Read how Financial AI in Practice explains model architecture →

Also covered in

What to know first

Related concepts

Questions about model architecture

What is model architecture?

Model architecture is the structural arrangement of a computational model’s components, connections, and operations that determines how it processes inputs and produces outputs.

Which Manning books explain model architecture?

Grokking Machine Learning, Delivering Machine Learning Projects: From design to deployment, Math and Architectures of Deep Learning, Lead Developer Career Guide, Architecture Modernization: Socio-technical alignment of software, strategy, and structure, Build AI Drug Discovery Pipelines, Timeless Algorithms, The Foundational Ideas, Functional Design and Architecture: Examples in Haskell and 14 more.

What should I know before learning model architecture?

Manning books that teach model architecture usually introduce language model, contract, machine learning and accuracy first.

What is the difference between model architecture and Apache Kafka?

Model architecture is the structural arrangement of a computational model’s components, connections, and operations that determines how it processes inputs and produces outputs. Apache Kafka is a distributed event-streaming platform for publishing, storing, and processing streams of records.

What kinds of model architecture are there?

Kinds of model architecture include transformer architecture, autoencoder, hexagonal architecture and linear layer.

What is model architecture also called?

Model architecture is also called neural network architecture, model’s architecture, deep learning architecture or neural architecture.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.