tff.tensorflow.computation

Decorates/wraps Python functions and defuns as TFF TensorFlow computations.

tff.tensorflow.computation(
    *args, **kwargs
)

Used in the notebooks

This symbol can be used as either a decorator or a wrapper applied to a function given to it as an argument. The supported patterns and examples of usage are as follows:

1. Convert an existing function inline into a TFF computation. This is the simplest mode of usage, and how one can embed existing non-TFF code for use with the TFF framework. In this mode, one invokes tff.tensorflow.computation with a pair of arguments, the first being a function/defun that contains the logic, and the second being the TFF type of the parameter:

   foo = tff.tensorflow.computation(lambda x: x > 10, tf.int32)
   

   After executing the above code snippet, foo becomes an instance of the abstract base class Computation. Like all computations, it has the type_signature property:

   str(foo.type_signature) == '(int32 -> bool)'
   

   The function passed as a parameter doesn't have to be a lambda, it can also be an existing Python function or a defun. Here's how to construct a computation from the standard TensorFlow operator tf.add:

   foo = tff.tensorflow.computation(tf.add, (tf.int32, tf.int32))
   

   The resulting type signature is as expected:

   str(foo.type_signature) == '(<int32,int32> -> int32)'
   

   If one intends to create a computation that doesn't accept any arguments, the type argument is simply omitted. The function must be a no-argument function as well:

   foo = tff.tensorflow.computation(lambda: tf.constant(10))
   

2. Decorate a Python function or a TensorFlow defun with a TFF type to wrap it as a TFF computation. The only difference between this mode of usage and the one mentioned above is that instead of passing the function/defun as an argument, tff.tensorflow.computation along with the optional type specifier is written above the function/defun's body.

   Here's an example of a computation that accepts a parameter:

   @tff.tensorflow.computation(tf.int32)
   def foo(x):
     return x > 10
   

   One can think of this mode of usage as merely a syntactic sugar for the example already given earlier:

   foo = tff.tensorflow.computation(lambda x: x > 10, tf.int32)
   

   Here's an example of a no-parameter computation:

   @tff.tensorflow.computation
   def foo():
     return tf.constant(10)
   

   Again, this is merely syntactic sugar for the example given earlier:

   foo = tff.tensorflow.computation(lambda: tf.constant(10))
   

   If the Python function has multiple decorators, tff.tensorflow.computation should be the outermost one (the one that appears first in the sequence).

3. Create a polymorphic callable to be instantiated based on arguments, similarly to TensorFlow defuns that have been defined without an input signature.

   This mode of usage is symmetric to those above. One simply omits the type specifier, and applies tff.tensorflow.computation as a decorator or wrapper to a function/defun that does expect parameters.

   Here's an example of wrapping a lambda as a polymorphic callable:

   foo = tff.tensorflow.computation(lambda x, y: x > y)
   

   The resulting foo can be used in the same ways as if it were had the type been declared; the corresponding computation is simply created on demand, in the same way as how polymorphic TensorFlow defuns create and cache concrete function definitions for each combination of argument types.

   ...foo(1, 2)...
   ...foo(0.5, 0.3)...
   

   Here's an example of creating a polymorphic callable via decorator:

   @tff.tensorflow.computation
   def foo(x, y):
     return x > y
   

   The syntax is symmetric to all examples already shown.