TensorFlow Dataset
TensorFlow Dataset is an iterable abstraction for representing, transforming, batching, and supplying sequences of data elements to machine-learning models.
Deep Learning with Python, Third Edition
TensorFlow's iterable dataset abstraction for efficient loading, preprocessing, and batching of data.
- How it is used
- Supplies iterable, preprocessed batches to Keras training, including image classification and distributed training.
- How deep it goes
- Works through directory loading, array construction, iteration, batching, and mapping, then gives a prefetching rule of thumb.
- Examples
- Uses directory-based image datasets and NumPy arrays to demonstrate creation, shape inspection, batching, mapping, and model fitting.
- What you can do after
- Apply it via `tf.data.Dataset` and Keras utilities to create training inputs and configure prefetching for distributed runs.
Read how Deep Learning with Python, Third Edition explains TensorFlow Dataset →
Machine Learning for Tabular Data: XGBoost, Deep Learning, and AI
tf.data.Dataset is a TensorFlow data abstraction representing a sequence of elements that can be transformed and iterated during model input processing.
- How it is used
- Builds efficient TensorFlow input pipelines for structured tabular model training, validation, and testing.
- How deep it goes
- Explains conversion, shuffling, batching, prefetching, and dataset splits, but defers Dataset internals.
- Examples
- Uses property data with categorical and continuous columns, then creates train, validation, and test datasets.
- What you can do after
- Builds a shuffled, batched, prefetched dataset from a pandas-style DataFrame for Keras model input.
Related concepts
- image_dataset_from_directoryproduced by · image_dataset_from_directory is a Keras utility function that creates an image dataset from files organized in class-specific subdirectories.
- tf.data APIuses · The tf.data API is TensorFlow’s API for constructing input pipelines that read, transform, batch, and iterate over data efficiently.
- fit methodused by · A fit method is an operation that learns or estimates model parameters, structure, or transformation statistics from supplied data.
- generationused by · Generation is the computational production of outputs from supplied inputs or context, often by applying a model to produce predictions or new content.
- prefetchused by · Prefetching is the proactive retrieval or preparation of anticipated data or resources before they are needed.
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.