Glossary term
Glossary term
Infrastructure and Serving
A TensorFlow programming environment in which operations run immediately. In contrast, operations called in graph execution don't run until they are explicitly evaluated. Eager execution is an imperative interface, much like the code in most programming languages. Eager execution programs are generally far easier to debug than graph execution programs.
Created for this library
A research team uses eager execution in TensorFlow during experimentation so each operation can be inspected immediately with standard Python debugging.
An ML platform team uses eager execution during model development and switches to graph execution for production serving to maximize throughput.
A startup's research engineer relies on eager execution for fast iteration on small models before committing to graph compilation for scale runs.
Definition source: Google for Developers Machine Learning Glossary | Creative Commons Attribution 4.0 License