• tf.contrib.eager.custom_gradient
  • tf.custom_gradient

Defined in tensorflow/python/ops/

Decorator to define a function with a custom gradient.

This decorator allows fine grained control over the gradients of a sequence for operations. This may be useful for multiple reasons, including providing a more efficient or numerically stable gradient for a sequence of operations.

For example, consider the following function that commonly occurs in the computation of cross entropy and log likelihoods:

def log1pexp(x):
  return tf.log(1 + tf.exp(x))

Due to numerical instability, the gradient this function evaluated at x=100 is NaN. For example:

x = tf.constant(100.)
y = log1pexp(x)
dy = tf.gradients(y, x) # Will be NaN when evaluated.

The gradient expression can be analytically simplified to provide numerical stability:

def log1pexp(x):
  e = tf.exp(x)
  def grad(dy):
    return dy * (1 - 1 / (1 + e))
  return tf.log(1 + e), grad

With this definition, the gradient at x=100 will be correctly evaluated as 1.0.

See also tf.RegisterGradient which registers a gradient function for a primitive TensorFlow operation. tf.custom_gradient on the other hand allows for fine grained control over the gradient computation of a sequence of operations.

Note that if the decorated function uses Variables, the enclosing variable scope must be using ResourceVariables.


  • f: function f(x) that returns a tuple (y, grad_fn) where:
    • x is a Tensor or sequence of Tensor inputs to the function.
    • y is a Tensor or sequence of Tensor outputs of applying TensorFlow operations in f to x.
    • grad_fn is a function with the signature g(*grad_ys) which returns a list of Tensors - the derivatives of Tensors in y with respect to the Tensors in x.grad_ysis aTensoror sequence ofTensors the same size asyholding the initial value gradients for eachTensoriny. IffusesVariables (that are not part of the inputs), i.e. throughget_variable, thengrad_fnshould have signatureg(*grad_ys, variables=None), wherevariablesis a list of theVariables, and return a 2-tuple(grad_xs, grad_vars), wheregrad_xsis the same as above, andgrad_varsis alistwith the derivatives ofTensors iny` with respect to the variables.


A function h(x) which returns the same value as f(x)[0] and whose gradient (as calculated by tf.gradients) is determined by f(x)[1].