Tune in to the first Women in ML Symposium this Tuesday, October 19 at 9am PST Register now

tensorflow:: ops:: IdentityN

#include <array_ops.h>

Returns a list of tensors with the same shapes and contents as the input.

Summary

tensors.

This op can be used to override the gradient for complicated functions. For example, suppose y = f(x) and we wish to apply a custom function g for backprop such that dx = g(dy). In Python,

with tf.get_default_graph().gradient_override_map(
    {'IdentityN': 'OverrideGradientWithG'}):
  y, _ = identity_n([f(x), x])

.RegisterGradient('OverrideGradientWithG')
def ApplyG(op, dy, _):
  return [None, g(dy)]  # Do not backprop to f(x).

Args:

Returns:

  • OutputList : The output tensor.

Constructors and Destructors

IdentityN (const :: tensorflow::Scope & scope, :: tensorflow::InputList input)

Public attributes

operation
output

Public functions

operator[] (size_t index) const

Public attributes

operation

Operation operation

output

::tensorflow::OutputList output

Public functions

IdentityN

 IdentityN(
  const ::tensorflow::Scope & scope,
  ::tensorflow::InputList input
)

operator[]

::tensorflow::Output operator[](
  size_t index
) const