tf.compat.v1.flags.DEFINE_enum | TensorFlow v2.16.1 (original) (raw)

tf.compat.v1.flags.DEFINE_enum

Stay organized with collections Save and categorize content based on your preferences.

Registers a flag whose value can be any string from enum_values.

tf.compat.v1.flags.DEFINE_enum(
    name,
    default,
    enum_values,
    help,
    flag_values=_flagvalues.FLAGS,
    module_name=None,
    required=False,
    **args
)

Instead of a string enum, prefer DEFINE_enum_class, which allows defining enums from an enum.Enum class.

Args
name	str, the flag name.
default	str\|None, the default value of the flag.
enum_values	[str], a non-empty list of strings with the possible values for the flag.
help	str, the help message.
flag_values	:class:FlagValues, the FlagValues instance with which the flag will be registered. This should almost never need to be overridden.
module_name	str, the name of the Python module declaring this flag. If not provided, it will be computed using the stack trace of this call.
required	bool, is this a required flag. This must be used as a keyword argument.
**args	dict, the extra keyword args that are passed to Flag.__init__.

Returns
a handle to defined flag.

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.