Save the date! Google I/O returns May 18-20 Register now


Computes rectified linear gradients for a LeakyRelu operation.

gradients A Tensor. Must be one of the following types: half, bfloat16, float32, float64. The backpropagated gradients to the corresponding LeakyRelu operation.
features A Tensor. Must have the same type as gradients. The features passed as input to the corresponding LeakyRelu operation, OR the outputs of that operation (both work equivalently).
alpha An optional float. Defaults to 0.2.
name A name for the operation (optional).

A Tensor. Has the same type as gradients.