/TensorFlow Python




Defined in tensorflow/python/util/deprecation.py.

See the guide: Framework (contrib) > Deprecation

Decorator for marking specific function arguments as deprecated.

This decorator logs a deprecation warning whenever the decorated function is called with the deprecated argument. It has the following format:

Calling (from ) with is deprecated and will be removed after . Instructions for updating:

If date is None, 'after ' is replaced with 'in a future version'. includes the class name if it is a method.

It also edits the docstring of the function: ' (deprecated arguments)' is appended to the first line of the docstring and a deprecation notice is prepended to the rest of the docstring.


  • date: String or None. The date the function is scheduled to be removed. Must be ISO 8601 (YYYY-MM-DD), or None.
  • instructions: String. Instructions on how to update code using the deprecated function. *deprecated_arg_names_or_tuples: String. or 2-Tuple(String, [ok_vals]). The string is the deprecated argument name. Optionally, an ok-value may be provided. If the user provided argument equals this value, the warning is suppressed.


Decorated function or method.


  • ValueError: If date is not None or in ISO 8601 format, instructions are empty, the deprecated arguments are not present in the function signature, or the second element of a deprecated_tuple is not a list.

© 2017 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 3.0.
Code samples licensed under the Apache 2.0 License.