TensorFlow
TensorFlow is an open-source software platform for numerical computation and machine learning, with tools for developing, training, and deploying models.
Deep Learning with R, Third Edition
TensorFlow is an open-source platform and software ecosystem for numerical computation and machine learning.
- How it is used
- Uses TensorFlow for low-level tensor computation and gradient-based training, either directly or as a Keras backend.
- How deep it goes
- Works through a classifier’s forward pass, loss, gradients, and updates, then shows a custom Keras training step without deriving automatic differentiation.
- Examples
- Uses a synthetic two-class linear classifier and a Keras model’s custom training step on MNIST.
- What you can do after
- Build it from scratch: train a simple TensorFlow classifier, or apply it as the backend for a custom Keras training step.
Read how Deep Learning with R, Third Edition explains TensorFlow →
Deep Learning with Python, Third Edition
A Python-based open source machine learning platform with an ecosystem of components for training, deployment, and related workflows.
- How it is used
- Provides low-level tensor operations and gradients for model training, alongside an ecosystem for preprocessing and deployment.
- How deep it goes
- Works through a linear classifier’s variables, forward pass, loss, gradients, and updates; surveys platform components without implementing deployment.
- Examples
- Builds a classifier for two synthetic 2D point classes and plots its predictions and separating line.
- What you can do after
- Build it from scratch: train a linear classifier with TensorFlow variables, tf.GradientTape, and repeated gradient updates.
Read how Deep Learning with Python, Third Edition explains TensorFlow →
Distributed Machine Learning Patterns
An open-source platform for numerical computation and machine learning, including tools for building and training neural networks.
- How it is used
- Builds, trains, and evaluates machine-learning models within an end-to-end distributed ML workflow.
- How deep it goes
- Works through installation, MNIST preprocessing, Sequential training, persistence, and hyperparameter tuning without covering distributed TensorFlow execution.
- Examples
- Demonstrates MNIST digit classification, model saving and loading, and Hyperband tuning with EarlyStopping.
- What you can do after
- Apply it via tf.keras to preprocess data, train and evaluate a local classifier, persist it, and tune selected hyperparameters.
Read how Distributed Machine Learning Patterns explains TensorFlow →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
TensorFlow is an open-source software platform for building and executing machine learning computations, including neural networks.
- How it is used
- Supports deep-learning workflows for tabular data beneath Keras and related libraries.
- How deep it goes
- Explains TensorFlow's stack role and contrasts APIs, but defers detailed layer definitions and framework internals.
- Examples
- Uses an Airbnb price classifier with categorical, continuous, and text inputs plus embedding and batch-normalization layers.
- What you can do after
- Evaluate TensorFlow with Keras as a valid option for deep learning with tabular data on Colab.
Read how Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI explains TensorFlow →
Related concepts
- Kerasused by · Keras is an open-source, high-level deep-learning API for building, training, evaluating, and deploying neural-network 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.
- PyTorchalternative to · PyTorch is an open-source machine-learning framework for Python that provides tensor computation, automatic differentiation, and tools for building and training neural networks.
- Pythonuses · Python is a high-level, general-purpose, dynamically typed programming language known for readable syntax and automatic memory management.
- backpropagationimplements · Backpropagation is a procedure for computing derivatives of an output or loss with respect to inputs and parameters by applying the chain rule backward through a computational graph.
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.