Torch.nn.Module

As explained in Inside Deep Learning: Math, Algorithms, Models

torch.nn.Module is PyTorch’s base class for neural-network components, providing a structure for computation and managing registered parameters and submodules.

A PyTorch Module organizes a neural-network component: its forward function maps inputs to outputs, while PyTorch handles backward computation unless you intervene. Standard modules live in torch.nn; custom modules can be built from tensors, Parameter, and torch.nn.functional objects, and can contain other modules. section 2.1.5

From layer to model

For d inputs and C outputs, nn.Linear(d, C) provides a linear layer with a bias. A larger model can compose layers, as in a Sequential classifier with Linear and Tanh layers, then pass the model to a training function. section 2.1.5 section 2.4.4

Figure 2.3 The Model box denotes the neural network f_Θ(∙) that we want to train. This could be a small network or a very complex one. It is encapsulated into one model object with a single set of parameters Θ. The same process works for almost any network definition.

Shows a neural network encapsulated as one model object with a single set of parameters. section 2.1.5

How Inside Deep Learning: Math, Algorithms, Models covers it

How deep it goes

Defines forward and automatic backward behavior and describes module composition, but does not detail custom-module implementation.

Examples

Uses a linear layer and a Sequential moons classifier built from Linear and Tanh layers.

What you can do after

Apply it via PyTorch by composing layers into a model and passing that model to a training function.

How it connects

Connects to nn.Linear, Parameter, torch.nn.functional, model.parameters(), and model.train()/eval().

liveBookmanning.com

Read Inside Deep Learning: Math, Algorithms, Models on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Deep Learning with PyTorch, Second EditionThe PyTorch base class for defining neural-network models as Python modules with parameters, submodules, and a forward method.
• 

  liveBookmanning.com
  Deep Learning with Python, Third EditionThe base class for defining stateful PyTorch models with parameters and a forward method.
• 

  liveBookmanning.com
  Natural Language Processing in Action, Second EditionA PyTorch module is a reusable neural-network component represented by the torch.nn.Module base class and its registered parameters and submodules.
• 

  liveBookmanning.com
  Math and Architectures of Deep LearningThe PyTorch base class that neural network models subclass to define learnable modules and a forward pass.
• 

  liveBookmanning.com
  Deep Learning with R, Third Editiontorch.nn.Module is PyTorch's base class for building objects that represent neural-network modules and manage their parameters.

Compare how 6 books explain torch.nn.Module →

Related concepts

• PyTorchused by · PyTorch is an open-source machine-learning framework that provides tensor computation, automatic differentiation, and tools for building and training neural networks.
• forwardimplements · In PyTorch, forward is the module method that defines how inputs are transformed into outputs during a forward pass.
• Sequentialused by · Sequential is a PyTorch container module that applies child modules in a specified order to form a composite computation.
• positional embeddingimplements · A positional embedding is a learned or fixed representation that encodes an element’s position or order in a sequence and is incorporated into its content representation.
• DataLoaderuses · PyTorch DataLoader is an iterable utility that retrieves dataset samples, optionally shuffles and transforms them, and groups them into batches.
• optimizationuses · Optimization is the systematic selection or adjustment of values, actions, or system designs to improve a specified objective, often subject to constraints.

This explainer was generated by Manning's knowledge graph from the text of Inside Deep Learning: Math, Algorithms, Models. The section links point to the parts of the book it draws on.