tf.compat.v1.global_variables | TensorFlow v2.16.1 (original) (raw)
tf.compat.v1.global_variables
Stay organized with collections Save and categorize content based on your preferences.
Returns global variables.
tf.compat.v1.global_variables(
scope=None
)
Migrate to TF2
Not compatible with eager execution and tf.function. In particular, Graph collections are deprecated in TF2. Instead please create atf.Modulecontainer for all your model state, including variables. You can then list all the variables in your tf.Module through thevariables attribute.
Description
Used in the notebooks
| Used in the guide |
|---|
| Validating correctness & numerical equivalence Migrating model checkpoints |
Global variables are variables that are shared across machines in a distributed environment. The Variable() constructor or get_variable()automatically adds new variables to the graph collectionGraphKeys.GLOBAL_VARIABLES. This convenience function returns the contents of that collection.
An alternative to global variables are local variables. Seetf.compat.v1.local_variables
| Args | |
|---|---|
| scope | (Optional.) A string. If supplied, the resulting list is filtered to include only items whose name attribute matches scope usingre.match. Items without a name attribute are never returned if a scope is supplied. The choice of re.match means that a scope without special tokens filters by prefix. |
| Returns |
|---|
| A list of Variable objects. |
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.
Last updated 2024-04-26 UTC.