By equivalent graph code we mean code that generates a TensorFlow graph when
run. The generated graph has the same effects as the original code when executed
(for example with tf.function or tf.compat.v1.Session.run). In other words,
using AutoGraph can be thought of as running Python in TensorFlow.
Modules
experimental module: Public API for tf.autograph.experimental namespace.
[null,null,["Last updated 2020-10-01 UTC."],[],[],null,["# Module: tf.autograph\n\n\u003cbr /\u003e\n\n|----------------------------------------------------------------------|\n| [TensorFlow 1 version](/versions/r1.15/api_docs/python/tf/autograph) |\n\nConversion of plain Python into TensorFlow graph code.\n| **Note:** In TensorFlow 2.0, AutoGraph is automatically applied when using [`tf.function`](../tf/function). This module contains lower-level APIs for advanced use.\n\nFor more information, see the\n[AutoGraph guide](https://www.tensorflow.org/guide/autograph).\n\nBy equivalent graph code we mean code that generates a TensorFlow graph when\nrun. The generated graph has the same effects as the original code when executed\n(for example with [`tf.function`](../tf/function) or [`tf.compat.v1.Session.run`](../tf/compat/v1/Session#run)). In other words,\nusing AutoGraph can be thought of as running Python in TensorFlow.\n\nModules\n-------\n\n[`experimental`](../tf/autograph/experimental) module: Public API for tf.autograph.experimental namespace.\n\nFunctions\n---------\n\n[`set_verbosity(...)`](../tf/autograph/set_verbosity): Sets the AutoGraph verbosity level.\n\n[`to_code(...)`](../tf/autograph/to_code): Returns the source code generated by AutoGraph, as a string.\n\n[`to_graph(...)`](../tf/autograph/to_graph): Converts a Python entity into a TensorFlow graph.\n\n[`trace(...)`](../tf/autograph/trace): Traces argument information at compilation time."]]
TensorFlow 1 version