Sequential model

As explained in Deep Learning with Python, Third Edition

A Keras class for building models as a linear stack of layers.

A Sequential model is the approachable Keras choice for a simple linear stack of layers: one input flows through each layer to one output. It is essentially a Python list of layers, so it works best when the model does not need branches, multiple inputs or outputs, or other graph-like connections. section 3.6.3 section 7.2

Build a small classifier

Create the stack directly, or add layers incrementally. This model transforms three input features through a 64-unit ReLU layer and produces ten softmax outputs: model = keras.Sequential([ layers.Dense(64, activation="relu"), layers.Dense(10, activation="softmax"), ]) The equivalent incremental form uses model.add(...), much like list.append(...). Give the model an input shape with keras.Input(shape=(3,)) when you want it built immediately and summarized as layers are added. section 7.2.1

Use the model-building and debugging workflow deliberately:

• Remember that layers create their weights only when the input shape is known. Calling the model on data or calling build(input_shape=...) builds it. section 7.2.1
• Call summary() after building to inspect output shapes and parameter counts; this is especially useful when layers transform inputs in complex ways. section 7.2.1
• Switch to the Functional API for directed acyclic graph architectures, or to model subclassing when you need low-level control. section 7.2 section 7.2.5

Figure 7.1 Progressive disclosure of complexity for model building

Use the complexity spectrum to compare Sequential models with the Functional API and model subclassing before choosing a construction style. section 7.2

How Deep Learning with Python, Third Edition covers it

How deep it goes

Explains direct and incremental construction, delayed weight creation, input-shape building, summaries, naming, and architecture limits.

Examples

Demonstrates Dense classifiers, incremental add calls, explicit input shapes, build(), weights, summary(), and named layers.

What it assumes

Assumes Python list-like usage and Keras layers while teaching input-dependent building and model inspection.

What you can do after

Build it from scratch with keras.Sequential, inspect its structure, and choose a more flexible API when the architecture outgrows a stack.

How it connects

Connects layer stacks with the Model class, Functional API, model subclassing, tensor operations, and network topology.

liveBookmanning.com

Read Deep Learning with Python, Third Edition on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Deep Learning with R, Third EditionA sequential model is a Keras model that represents a simple linear stack of layers with one input and one output.

Compare how 2 books explain Sequential model →

Related concepts

• Functional APIalternative to · The Functional API is a Keras interface for constructing neural-network models as directed computation graphs by connecting layers and tensors.
• Keras modelkind of · A Keras model is a trainable object representing a neural-network computation and providing interfaces for training, evaluation, prediction, and saving.
• model subclassingalternative to · Model subclassing is a technique for defining a neural-network model by creating a class that extends a model base class and specifies its computation.
• listequivalent to · A list is an ordered collection of elements arranged in a sequence.

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with Python, Third Edition. The section links point to the parts of the book it draws on.