Aliases:
tf.contrib.eager.py_func
tf.py_function
tf.py_function(
func,
inp,
Tout,
name=None
)
Defined in tensorflow/python/ops/script_ops.py
.
Wraps a python function into a TensorFlow op that executes it eagerly.
This function allows expressing computations in a TensorFlow graph as
Python functions. In particular, it wraps a Python function func
in a once-differentiable TensorFlow operation that executes it with eager
execution enabled. As a consequence, tf.contrib.eager.py_func
makes it
possible to express control flow using Python constructs (if
, while
,
for
, etc.), instead of TensorFlow control flow constructs (tf.cond
,
tf.while_loop
). For example, you might use tf.contrib.eager.py_func
to
implement the log huber function:
def log_huber(x, m):
if tf.abs(x) <= m:
return x**2
else:
return m**2 * (1 - 2 * tf.log(m) + tf.log(x**2))
x = tf.placeholder(tf.float32)
m = tf.placeholder(tf.float32)
y = tf.contrib.eager.py_func(func=log_huber, inp=[x, m], Tout=tf.float32)
dy_dx = tf.gradients(y, x)[0]
with tf.Session() as sess:
# The session executes `log_huber` eagerly. Given the feed values below,
# it will take the first branch, so `y` evaluates to 1.0 and
# `dy_dx` evaluates to 2.0.
y, dy_dx = sess.run([y, dy_dx], feed_dict={x: 1.0, m: 2.0})
You can also use tf.contrib.eager.py_func
to debug your models at runtime
using Python tools, i.e., you can isolate portions of your code that
you want to debug, wrap them in Python functions and insert pdb
tracepoints
or print statements as desired, and wrap those functions in
tf.contrib.eager.py_func
.
For more information on eager execution, see the Eager guide.
tf.contrib.eager.py_func
is similar in spirit to tf.py_func
, but unlike
the latter, the former lets you use TensorFlow operations in the wrapped
Python function. In particular, while tf.py_func
only runs on CPUs and
wraps functions that take NumPy arrays as inputs and return NumPy arrays as
outputs, tf.contrib.eager.py_func
can be placed on GPUs and wraps functions
that take Tensors as inputs, execute TensorFlow operations in their bodies,
and return Tensors as outputs.
Like tf.py_func
, tf.contrib.eager.py_func
has the following limitations
with respect to serialization and distribution:
The body of the function (i.e.
func
) will not be serialized in aGraphDef
. Therefore, you should not use this function if you need to serialize your model and restore it in a different environment.The operation must run in the same address space as the Python program that calls
tf.contrib.eager.py_func()
. If you are using distributed TensorFlow, you must run atf.train.Server
in the same process as the program that callstf.contrib.eager.py_func()
and you must pin the created operation to a device in that server (e.g. usingwith tf.device():
).
Args:
func
: A Python function which accepts a list ofTensor
objects having element types that match the correspondingtf.Tensor
objects ininp
and returns a list ofTensor
objects (or a singleTensor
, orNone
) having element types that match the corresponding values inTout
.inp
: A list ofTensor
objects.Tout
: A list or tuple of tensorflow data types or a single tensorflow data type if there is only one, indicating whatfunc
returns; an empty list if no value is returned (i.e., if the return value isNone
).name
: A name for the operation (optional).
Returns:
A list of Tensor
or a single Tensor
which func
computes; an empty list
if func
returns None.