tf.keras is TensorFlow's high-level API for building and training deep
learning models. It's used for fast prototyping, state-of-the-art research,
and production, with three key advantages:
- User-friendly
Keras has a simple, consistent interface optimized for common use cases. It provides clear and actionable feedback for user errors. - Modular and composable
Keras models are made by connecting configurable building blocks together, with few restrictions. - Easy to extend
Write custom building blocks to express new ideas for research. Create new layers, metrics, loss functions, and develop state-of-the-art models.
The guide Keras: A Quick Overview will help you get started.
For a beginner-friendly introduction to machine learning with tf.keras,
see this set of starter tutorials.
To dive more into the API, see the following set of guides that cover what you need to know as a TensorFlow Keras power user:
- Guide to the Keras functional API
- Guide to training and evaluation
- Guide to writing layers and models from scratch with subclassing
- Guide to the recurrent neural network API
- Guide to masking and padding
- Guide to saving and serializing models
- Guide to writing custom callbacks
Watch Inside TensorFlow on YouTube for a Keras internals deep dive: