Converts a Python entity into a TensorFlow graph.
tf.autograph.to_graph( entity, recursive=True, experimental_optional_features=None )
to_graph is a low-level transpiler that converts
Python code to TensorFlow graph code. It does not implement any caching,
variable management or create any actual ops, and is best used where greater
control over the generated TensorFlow graph is desired. Another difference
tf.function is that
to_graph will not wrap the graph into a
TensorFlow function or a Python callable. Internally,
def foo(x): if x > 0: y = x * x else: y = -x return y converted_foo = to_graph(foo) x = tf.constant(1) y = converted_foo(x) # converted_foo is a TensorFlow Op-like. assert is_tensor(y)
Supported Python entities include:
- object methods
Functions are converted into new functions with converted code.
Classes are converted by generating a new class whose methods use converted code.
Methods are converted into unbound function that have an additional first
entity: Python callable or class to convert.
recursive: Whether to recursively convert any functions that the converted function may call.
None, a tuple of, or a single
tf.autograph.experimental.Featurevalue. Controls the use of optional features in the conversion process.
entity, the converted Python function or class.
ValueError: If the entity could not be converted.