Keras
Keras is an open-source, high-level deep-learning API for building, training, and evaluating neural-network models.
Deep Learning with R, Third Edition
Keras is an open-source deep learning API for building and training neural networks.
- How it is used
- Provides a high-level interface for building, training, evaluating, and using neural networks.
- How deep it goes
- Progresses from a complete classifier to backend configuration, custom training steps, and distributed training.
- Examples
- Uses MNIST handwritten-digit classification and custom training steps for TensorFlow, PyTorch, and JAX.
- What you can do after
- Apply Keras via the R keras3 package to train, evaluate, and use a neural network for predictions.
Read how Deep Learning with R, Third Edition explains Keras →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
Keras is an open-source deep learning API for Python that provides interfaces for building and training neural-network models.
- How it is used
- Implements deep-learning solutions for tabular price-prediction and classification problems.
- How deep it goes
- Explains model layers, fit arguments, callbacks, persistence, evaluation, and repeated-run validation without detailing TensorFlow internals.
- Examples
- Uses Airbnb NYC listing-price prediction and Kuala Lumpur real-estate price classification.
- What you can do after
- Apply it via Keras to define, train, checkpoint, load, and evaluate tabular deep-learning models.
Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains Keras →
Deep Learning with Python, Third Edition
The programming interface of Keras that exposes multiple model-construction workflows and lets them interoperate.
- How it is used
- Provides a high-level interface for assembling, training, and evaluating neural networks across workflows with different flexibility.
- How deep it goes
- Explains model training and custom layer construction, then describes interoperable workflows spanning high-level fit() to custom logic.
- Examples
- Uses MNIST digit classification and a custom dense layer to demonstrate model and layer APIs.
- What you can do after
- Apply it via Keras to construct, train, and evaluate a digit classifier and configure a backend.
Read how Deep Learning with Python, Third Edition explains Keras →
Grokking Machine Learning
An open-source deep-learning library providing high-level APIs for constructing and training neural networks.
- How it is used
- Uses Keras to put neural-network theory into practice through classification examples.
- How deep it goes
- Demonstrates preprocessing, layer construction, compilation, training, and prediction without detailing library internals.
- Examples
- Uses a two-feature circular dataset and MNIST handwritten digits to demonstrate classification.
- What you can do after
- Build from scratch and train classifiers for nonlinear points and handwritten digits.
Distributed Machine Learning Patterns
Keras is a high-level deep-learning API for defining, training, evaluating, and exporting neural-network models, commonly used through TensorFlow as tf.keras.
- How it is used
- Uses a high-level TensorFlow API to move from dataset loading through neural-network training, evaluation, and persistence.
- How deep it goes
- Defines layers, compilation, fitting, evaluation, saving, and tuning, but does not descend into TensorFlow internals.
- Examples
- Demonstrates normalized MNIST images, a Sequential classifier, model persistence, and Keras Tuner Hyperband with early stopping.
- What you can do after
- Build it from scratch: create, train, evaluate, save, reload, and tune a TensorFlow neural-network model.
Read how Distributed Machine Learning Patterns explains Keras →
Evolutionary Deep Learning: Genetic algorithms and neural networks
Keras is an open-source deep-learning library providing high-level APIs for constructing and training neural networks.
- How it is used
- Deploys high-level neural-network construction and training for classification, optimizer comparison, and neuroevolution experiments.
- How deep it goes
- Descends through executable model definition, compilation, fitting, parameter counting, and weight replacement, but not framework internals.
- Examples
- Uses moons and circles classification datasets, Dense-layer models, loss plots, prediction displays, and evolved model weights.
- What you can do after
- Builds and trains a Keras classifier, inspects its parameters, and integrates genetic-algorithm weight optimization.
Read how Evolutionary Deep Learning: Genetic algorithms and neural networks explains Keras →
Fight Fraud with Machine Learning
Keras is a high-level deep-learning API for constructing and training neural networks, integrated into TensorFlow as tf.keras.
- How it is used
- Builds and trains deep neural networks through TensorFlow’s official high-level API.
- How deep it goes
- Explains architecture, compilation, and model inspection through runnable code without descending into framework internals.
- Examples
- Uses a Sequential sigmoid network with three hidden layers, Adam optimization, binary cross-entropy, accuracy, and a 271-parameter summary.
- What you can do after
- Build it from scratch with tf.keras by defining layers, compiling training settings, and inspecting the resulting model.
Also covered in
- Data Without Labels
- Deep Learning with PyTorch, Second Edition
- Grokking Machine Learning, Second EditionIn early access (MEAP)
What to know first
- accuracyAccuracy is the degree to which a result agrees with a specified truth, reference, or criterion, commonly measured as the proportion of evaluated predictions that are correct.
- deep learningDeep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
- PythonPython is a high-level, general-purpose, dynamically typed programming language with readable syntax and support for multiple programming paradigms.
- parameterA parameter is a value or named input that specifies or influences the behavior, computation, or output of a function, model, or system.
Related concepts
- TensorFlowuses · TensorFlow is an open-source software platform for numerical computation and machine learning, supporting the development, training, and execution of computational models.
- autoencoderused by · An autoencoder is a neural network that encodes input data into an internal representation and decodes it to reconstruct the input or an approximation.
- deep learninguses · Deep learning is a machine-learning approach that uses neural networks with multiple computational layers to learn representations from data.
- backenduses · A backend is a software component or system that provides underlying processing, data, execution, or service functionality to other components.
- backpropagationimplements · Backpropagation is reverse-mode differentiation that propagates derivatives backward through a computational graph to compute gradients of an output or loss with respect to inputs and parameters.
- JAXuses · JAX is a Python library for high-performance array computing, with composable transformations for automatic differentiation, compilation, vectorization, and parallelization.
Questions about Keras
What is Keras?
Keras is an open-source, high-level deep-learning API for building, training, and evaluating neural-network models.
Which Manning books explain Keras?
Deep Learning with R, Third Edition, Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI, Deep Learning with Python, Third Edition, Grokking Machine Learning, Distributed Machine Learning Patterns, Evolutionary Deep Learning: Genetic algorithms and neural networks and Fight Fraud with Machine Learning.
What should I know before learning Keras?
Manning books that teach Keras usually introduce accuracy, deep learning, Python and parameter first.
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.






