TensorFlow

API

 tf.compat / v1 / v1.flags.MultiFlag


A function decorator for defining a multi-flag validator.

Registers the decorated function as a validator for flag_names, e.g.

@flags.multi_flags_validator(['foo', 'bar']) def _CheckFooBar(flags_dict): ...

See register_multi_flags_validator() for the specification of checker function.

flag_names [str], a list of the flag names to be checked.
message str, error text to be shown to the user if checker returns False. If checker raises flags.ValidationError, message from the raised error will be shown.
flag_values flags.FlagValues, optional FlagValues instance to validate against.

A function decorator that registers its function argument as a validator.

AttributeError Raised when a flag is not registered as a valid flag name.

此页内容是否对您有帮助