TensorFlow 2 focuses on simplicity and ease of use, with updates like eager execution, intuitive higher-level APIs, and flexible model building on any platform.

Many guides are written as Jupyter notebooks and run directly in Google Colab—a hosted notebook environment that requires no setup. Click the Run in Google Colab button.

Essential documentation

Install the package or build from source. GPU support for CUDA®-enabled cards.
TensorFlow 2 best practices and tools to migrate your code.
Keras is a high-level API that's easier for ML beginners, as well as researchers.
TensorFlow for research and experimentation. Write custom layers and models, forward pass, and training loops.
The tf.data API enables you to build complex input pipelines from simple, reusable pieces.
A high-level API that represents a complete model, designed for scaling and asynchronous training.
Save a TensorFlow model using checkpoints or the SavedModel format.
Distribute training across multiple GPUs, multiple machines or TPUs.
Explore additional resources to build advanced models or methods using TensorFlow, and access domain-specific application packages that extend TensorFlow.