tf.contrib.autograph.ConversionOptions

Class ConversionOptions

Defined in tensorflow/python/autograph/impl/api.py.

Container for conversion flags.

Attributes:

  • recursive: bool, whether to recursively convert any user functions or classes that the converted function may use.
  • verbose: bool, whether to log the compiled code.
  • strip_decorators: Tuple[Callable], contains decorators that should be in excluded from the compiled output. By default, when converting a function before the decorators are applied, the compiled output will include those decorators.
  • force_conversion: bool, whether to force convertinng the target entity. When force_conversion is turned off, the converter may decide to return the function as-is.
  • arg_types: Optional[Dict[Text, Type]], type hints for symbols including function arguments.

__new__

__new__(
    _cls,
    recursive,
    verbose,
    strip_decorators,
    force_conversion,
    arg_types
)

Create new instance of ConversionOptions(recursive, verbose, strip_decorators, force_conversion, arg_types)

Properties

recursive

verbose

strip_decorators

force_conversion

arg_types

Methods

new

@classmethod
new(
    cls,
    recursive=False,
    verbose=False,
    strip_decorators=None,
    force_conversion=False,
    arg_types=None
)