15d9064b86
* Create rule S6908 * Create rule S6908: "tf.function" should not be recursive. --------- Co-authored-by: joke1196 <joke1196@users.noreply.github.com> Co-authored-by: David Kunzmann <david.kunzmann@sonarsource.com>
54 lines
2.1 KiB
Plaintext
54 lines
2.1 KiB
Plaintext
This rule raises an issue when a `tensorflow.function` is recursive.
|
|
|
|
== Why is this an issue?
|
|
|
|
When defining a `tensorflow.function` it is generally a bad practice to make this function recursive. TensorFlow does not
|
|
support recursive `tensorflow.function` and will in the majority of cases throw an exception. However it is possible as
|
|
well that the execution of such function succeeds, but with multiple tracings which has strong performance implications.
|
|
When executing `tensorflow.function`, the code is split into two distinct stages.
|
|
The first stage call `tracing` creates a new `tensorflow.Graph`, runs the Python code normally,
|
|
but defers the execution of TensorFlow operations (i.e. adding two Tensors). These operations are added to the graph without being ran.
|
|
The second stage which is much faster than the first, runs everything that was deferred previously.
|
|
Depending on the input of the `tensorflow.function` the first stage may not be needed, see: https://www.tensorflow.org/guide/function#rules_of_tracing[Rules of tracing].
|
|
Skipping this first stage is what provides the user with TensorFlow's high performance.
|
|
|
|
Having a recursive `tensorflow.function` prevents the user from benefiting of TensorFlow's capabilities.
|
|
|
|
== How to fix it
|
|
|
|
To fix this issue, refactor the `tensorflow.function` so that is it not recursive.
|
|
|
|
=== Code examples
|
|
|
|
==== Noncompliant code example
|
|
|
|
[source,python,diff-id=1,diff-type=noncompliant]
|
|
----
|
|
import tensorflow as tf
|
|
|
|
@tf.function
|
|
def factorial(n):
|
|
if n == 1:
|
|
return 1
|
|
else:
|
|
return (n * factorial(n-1)) # Noncompliant: the function is recursive
|
|
----
|
|
|
|
==== Compliant solution
|
|
|
|
[source,python,diff-id=1,diff-type=compliant]
|
|
----
|
|
import tensorflow as tf
|
|
|
|
@tf.function
|
|
def factorial(n):
|
|
return tf.exp(tf.lgamma(n + 1)) # Compliant
|
|
----
|
|
|
|
|
|
== Resources
|
|
=== Documentation
|
|
|
|
* TensorFlow Documentation - https://www.tensorflow.org/guide/function#recursive_tffunctions_are_not_supported[Recursive tf.functions are not supported]
|
|
* TensorFlow Documentation - https://www.tensorflow.org/guide/function#rules_of_tracing[Rules of tracing]
|